Вакансии

Location: Tel Aviv-Yafo, Tel Aviv District, Israel About Aidoc Aidoc is the market leader in healthcare AI and has developed the world's largest clinical frontier model, supporting nearly 50 million patients each year. Aidoc is deployed in more than 2,000 medical centers worldwide. Since our founding in 2016, Aidoc has raised over $500 million and achieved a record number of FDA-cleared solutions. Medical diagnosis is hitting a breaking point, where the number of physicians hasn't kept up with the rapidly growing patient load. This leads to misdiagnosis and treatment delays that result in hundreds of thousands of deaths every year. Aidoc assists clinicians by precisely highlighting diseases directly on medical images, preventing misdiagnosis and ensuring that urgent patients receive the immediate attention they need. About this role We’re looking for a Principal AI Software Engineer to lead the design and development of the evolution and scaling of Aidoc’s AI architecture. You’ll combine hands-on development with architectural leadership - solving complex engineering challenges, guiding your team through design decisions, and ensuring our systems are robust, scalable, and maintainable. You’ll collaborate with other senior engineers, tech leads, and product partners to define technical direction, drive innovation, and improve engineering practices across the teams and groups. This is a role for someone who thrives on building well-designed systems, driving technical excellence within the team and mentoring peers. Responsibilities Lead the design and implementation of scalable and reliable backend services. Drive architectural discussions and make key technical decisions within your domain. Collaborate with algo teams and cross-functional teams to translate business and product goals into technical solutions. Deliver high-quality, maintainable, and testable code while ensuring best practices for security, performance, and reliability. Mentor and support engineers through design reviews, code reviews, and knowledge sharing. Continuously improve development processes, observability, and deployment pipelines. Contribute to the evolution of the platform’s architecture as the company and product scale. Requirements BA/BSc/MSc in Computer Science, Engineering or a related field. 10+ years of hands-on experience designing and developing complex, production-grade systems. Proven track record in architecting and leading large-scale software projects. Deep understanding of system design, distributed systems, and software architecture patterns. Strong coding skills in one or more modern backend languages (e.g., Python, Go, Node.js, Java, C#). Experience with cloud environments (AWS or equivalent), containerized deployments, and infrastructure as code (e.g., Docker, Kubernetes, Terraform). Experience designing and scaling event-driven architectures and data-intensive systems. Excellent communication and influencing skills - able to align multiple teams toward a shared technical vision. A proactive mindset, strong sense of ownership, and passion for solving complex engineering problems. Advantage: Experience in developing AI and machine learning solutions, with a focus on computer vision and NLP. Working at Aidoc Our Perks: Be part of something big - using cutting-edge technologies to transform the Healthcare industry (while saving patients’ lives) We work in a hybrid model, with our new offices located at 34 HaMasger Street in Tel Aviv and parking for employees. Amazing and healthy breakfasts and lunches prepared daily by our personal chef! Stocked up kitchen & meal card Wellness: Aidoc employees-only gym, plus Pilates, Yoga, and functional workouts classes. Amazing culture - collaborative, transparent & fun! Attractive compensation package & benefits Aidoc is deeply committed to creating an inclusive workplace, and to the principle of equal opportunity for all individuals. We prohibit discrimination and harassment based on race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, or any other status protected by law.

Location: Europe Workplace: Remote Synthesia is the world’s leading AI video platform for business, used by over 90% of the Fortune 100. Founded in 2017, the company is headquartered in London, with offices and teams across Europe and the US. As AI continues to shape the way we live and work, Synthesia develops products to enhance visual communication and enterprise skill development, helping people work better and stay at the center of successful organizations. Following our recent Series E funding round, where we raised $200 million, our valuation stands at $4 billion. Our total funding exceeds $530 million from premier investors including Accel, NVentures (Nvidia's VC arm), Kleiner Perkins, GV, and Evantic Capital, alongside the founders and operators of Stripe, Datadog, Miro, and Webflow. We’re looking for a Principal Engineer to join the ML Platform team at Synthesia. Our team builds and operates the systems that allow researchers and product teams to train, serve, and deploy generative models reliably and efficiently. This includes research infrastructure, production serving systems, internal tooling, and the platform interfaces that connect them. A growing part of our mission is making these systems more automation-friendly and agent-oriented, so that workflows can increasingly be operated through reliable tooling rather than manual effort. We’re looking for a strong generalist with a systems mindset: - someone who is comfortable working across infrastructure, backend systems, and tooling, and who has seen ML systems in practice. - this is not a pure ML Engineer role. We’re especially interested in people who think deeply about reliability, scalability, performance, and resource efficiency in complex production environments. This is a hands-on IC role with significant ownership. You’ll help shape how our ML platform evolves as we scale the number of models, workloads, tools and teams relying on it. WHAT YOU’LL DO - Design and improve the platform systems that support model training, evaluation, and production serving. - Build infrastructure and tooling that make ML workloads more reliable, scalable, and cost-efficient. - Develop internal tools and workflows that are easy to operate both by humans and by agents. - Work on the architecture behind how models are deployed, served, and operated across research and product environments. - Improve how we schedule, monitor, and debug workloads running on GPUs and cloud infrastructure. - Develop internal tools and abstractions and agentic systems that reduce operational overhead for researchers and engineers. - Drive improvements across observability, automation, reliability, and developer experience. - Collaborate closely with researchers and product engineers to understand pain points and turn them into robust platform capabilities. - Contribute to technical direction and make pragmatic architectural tradeoffs as the platform grows. YOU’LL THRIVE IN THIS ROLE IF YOU HAVE - Strong experience building or operating production systems with a focus on reliability, scalability, and maintainability. - A systems mindset: you naturally think in terms of bottlenecks, failure modes, interfaces, resource usage, and long-term operability. - Solid hands-on experience with cloud infrastructure, Linux, and infrastructure automation. - Experience with Kubernetes and operating distributed workloads in production. - Strong coding skills, ideally in Python or similar languages used for backend systems and tooling. - Strong judgment around where automation adds leverage, and where human control and reliability matter most. - Experience building internal platforms, developer tooling, or infrastructure abstractions used by other engineers. - Comfort working in ambiguous environments and taking ownership of open-ended technical problems. - A pragmatic approach: you care about solving the right problem well, not over-engineering. PARTICULARLY RELEVANT EXPERIENCE - Operating ML infrastructure or model serving systems in production. - Supporting research or data-intensive workloads. - Working with GPU-based systems or other performance-sensitive infrastructure. - Experience with observability and debugging in distributed systems. - Familiarity with Terraform, Datadog, GitHub Actions, or similar tools. BONUS POINTS FOR - Experience building agentic or LLM-powered internal tools. - Experience with workflow orchestration systems such as Temporal. - Experience working at the boundary between research and production engineering. - Familiarity with performance optimization, scheduling, or resource allocation problems. - Experience building lightweight product or developer-facing tools.

Задачи: — Проектирование и реализация бэкенд-микросервисов и API, интегрируемых с фундаментальными моделями для обработки документов и чертежей. — Построение масштабируемых стратегий маршрутизации запросов, кэширования и очереди для вывода, способных выдерживать резкие всплески нагрузки. — Реализация CI/CD и механизмов наблюдения за состоянием конвейеров вывода (логирование, метрики, оповещения). — Развёртывание и эксплуатация сервисов в Kubernetes (ответственность за развёртывание, выкатку и проверку здоровья; администрирование кластера обеспечивается платформой/SRE). — Интеграция сервисов с услугами AWS и обеспечение правильной идентификации доступа (IAM) и наименьшей привилегии для межсервисного взаимодействия. — Сотрудничество с учёными по данным и инженерами машинного обучения (они отвечают за обучение моделей; вы за интеграцию в производство). — Формирование производственных практик эксплуатации вывода, включая поддержку инцидентов на уровне службы. Требования: — Минимум 5 лет опыта создания промышленных бэкенд-систем и API на Python. — Опыт проектирования и эксплуатации распределённых микросервисов (масштабируемость, маршрутизация, кэширование, режимы отказов). — Практический опыт работы с AWS и интеграции внешних API моделей (преимущественно через Bedrock). — Опыт промышленной эксплуатации сервисов в Kubernetes (ответственность за развёртывание и эксплуатацию приложений; инфраструктура кластера поддерживается платформой). — Опыт работы с CI/CD, знание GitHub Actions. — Твердые практические знания по IAM и основным принципам безопасности в облаке. — Понимание логирования и наблюдаемости, умение проектировать конвейеры для анализа. — Опыт работы с Terraform или иными инструментами инфраструктуры как кода (IaC). — Способность ясно и убедительно передавать решения по проектированию систем и предыдущие проекты. Будет плюсом: — Опыт интеграции GenAI и LLMs (Bedrock, маршрутизация API, выбор моделей). — Опыт работы с конвейерами обработки изображений и документов (OCR, предобработка изображений, распознавание технических чертежей). — Опыт с паттернами оркестровки выводов (асинхронные работники, батчинг, оркестровка GPU) или знакомство с нагрузками на GPU. — Знание инструментов жизненного цикла ML (MLflow, регистры моделей, трекинг экспериментов). — Опыт работы с сервисами реального времени и пакетной обработки выводов, хранилищами признаков или производственной средой SageMaker. Условия: — Удалённая работа. — Полная занятость, проектная работа до конца года с возможным продлением. — Работа в англоговорящей команде, часовой пояс Германии.

Location: San Francisco, CA | New York City, NY | Seattle, WA About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems. About the role: Anthropic’s performance and scaling teams focus on making the most efficient and impactful use of our compute resources, be it inference or training.  As an Engineering Manager on these teams you will be responsible for ensuring you and your team are identifying and removing bottlenecks, building robust and durable solutions, and maximizing the efficiency of our systems.  You also will help bring clarity, focus, and context to your teams in a fast paced, dynamic environment. Responsibilities: Provide front-line leadership of engineering efforts to improve model performance and scale our inference and training systems Become familiar with the team’s technical stack enough to make targeted contributions as an individual contributor Manage day-to-day execution of the team's work Prioritize the team’s work and manage projects in a highly dynamic, fast paced environment Coach and support your reports in understanding, and pursuing, their professional growth Maintain a deep understanding of the team's technical work and its implications for AI safety You may be a good fit if you: Have 1+ years of management experience in a technical environment, particularly performance or distributed systems Have a background in machine learning, AI, or a similar related technical field Are deeply interested in the potential transformative effects of advanced AI systems and are committed to ensuring their safe development Excel at building strong relationships with stakeholders at all levels Are a quick learner, capable of understanding and contributing to discussions on complex technical topics Have experience managing teams through periods of rapid growth and change Are a quick study: this team sits at the intersection of a large number of different complex technical systems that you’ll need to understand (at a high level of abstraction) to be effective Strong candidates may also have experience with: High performance, large-scale ML systems GPU/Accelerator programming ML framework internals OS internals Language modeling with transformers The annual compensation range for this role is listed below. For sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role. Annual Salary: $500,000 — $850,000 USD Logistics Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience Required field of study: A field relevant to the role as demonstrated through coursework, training, or professional experience Minimum years of experience: Years of experience required will correlate with the internal job level requirements for the position Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices. Visa sponsorship: We do sponsor visas! However, we aren't able to successfully sponsor visas for every role and every candidate. But if we make you an offer, we will make every reasonable effort to get you a visa, and we retain an immigration lawyer to help with this. For roles performing work in California: We will consider for employment all qualified applicants, including those with arrest or conviction records or criminal histories, in a manner consistent with the requirements of applicable state and local laws, including the San Francisco Fair Chance Ordinance, the City of Los Angeles’ Fair Chance Initiative for Hiring Ordinance, the Los Angeles County Fair Chance Ordinance for Employers, and the California Fair Chance Act. We encourage you to apply even if you do not believe you meet every single qualification. Not all strong candidates will meet every single qualification as listed.  Research shows that people who identify as being from underrepresented groups are more prone to experiencing imposter syndrome and doubting the strength of their candidacy, so we urge you not to exclude yourself prematurely and to submit an application if you're interested in this work. We think AI systems like the ones we're building have enormous social and ethical implications. We think this makes representation even more important, and we strive to include a range of diverse perspectives on our team. Your safety matters to us. To protect yourself from potential scams, remember that Anthropic recruiters only contact you from @anthropic.com email addresses. In some cases, we may partner with vetted recruiting agencies who will identify themselves as working on behalf of Anthropic. Be cautious of emails from other domains. Legitimate Anthropic recruiters will never ask for money, fees, or banking information before your first day. If you're ever unsure about a communication, don't click any links—visit anthropic.com/careers directly for confirmed position openings. How we're different We believe that the highest-impact AI research will be big science. At Anthropic we work as a single cohesive team on just a few large-scale research efforts. And we value impact — advancing our long-term goals of steerable, trustworthy AI — rather than work on smaller and more specific puzzles. We view AI research as an empirical science, which has as much in common with physics and biology as with traditional efforts in computer science. We're an extremely collaborative group, and we host frequent research discussions to ensure that we are pursuing the highest-impact work at any given time. As such, we greatly value communication skills. The easiest way to understand our research directions is to read our recent research. This research continues many of the directions our team worked on prior to Anthropic, including: GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences. Come work with us! Anthropic is a public benefit corporation headquartered in San Francisco. We offer competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a lovely office space in which to collaborate with colleagues. Guidance on Candidates' AI Usage: Learn about our policy for using AI in our application process.

Разработка, обучение и оптимизация моделей машинного обучения для решения прикладных бизнес-задач; Разработка решений на базе LLM и VLM, включая мультимодальные системы обработки текста и изображений;

Location: Hamburg Workplace: Hybrid adjoe builds the technologies behind mobile apps growth and monetization. With our core product Playtime Arcade, we've become the global leader in rewarded advertising, an ad unit built on a simple premise: users earn real in-app rewards for engaging with new apps. The result is one of the most effective value exchanges in adtech, connecting advertisers and publishers with over 770 million users annually. Architecting Intelligence to Optimize 200M+ Daily Decisions As the intelligence core of our engineering organization, our Data Science team doesn't just deploy models, we engineer the fundamental decision engine that powers our platform. At a scale of 770 million users and 100,000+ predictions per second, we are solving a multi-objective optimization problem that balances user incentives, advertiser ROI, and long-term platform health in real time. Our architecture is built on a 1PB+ behavioral data lake, providing the high-fidelity input necessary to train deep learning models that predict individual user engagement with precision. We aren't just optimizing clicks, we are dynamically calculating optimal reward structures to sustain a global value exchange. Engineered for performance, our stack leverages PyTorch for model training, NVIDIA Triton to achieve sub-100ms inference. We own the full ML lifecycle from high-level research and feature engineering to deployment and A/B experimentation. Here, you will find the autonomy, the data depth, and the massive scale required to solve the most complex optimization challenges in the adtech ecosystem. The data infrastructure behind this runs at real scale: 2TB ingested in real time every day, 100+ Airflow jobs and data pipelines, and ML models handling p99 latency of 100ms across 100,000+ predictions per second. Your Mission & Who We Are Looking For - Quantitative Academic Foundation: You are currently pursuing a Bachelor’s or Master’s in Data Science, Computer Science, Mathematics, Statistics, or a related quantitative field. You are available to contribute up to 20 hours per week during the semester. - ML Model Prototyping: You will support the research and evaluation of deep learning and machine learning models that predict user behavior across our 770 million user base. We are looking for solid Python skills and hands-on familiarity with PyTorch, TensorFlow, scikit-learn, and gradient boosting libraries (LightGBM/CatBoost/XGBoost) to prototype new algorithms. - Large-Scale Data Exploration: Operating on our 1PB+ behavioral data lake, you will extract patterns from terabytes of data to engineer high-fidelity features for our training pipelines. This requires strong SQL fundamentals, an interest in distributed query engines (AWS Athena, Trino, Spark), and fluency with pandas, polars, and NumPy. - Experimentation & Validation: You will collaborate closely with our data scientists to test your fit-for-purpose solutions against our multi-objective optimization problem. We need someone who can rigorously evaluate models through offline experiments and A/B tests before they hit our 100,000+ predictions-per-second production environment. - Plus: ML Infrastructure Exposure: Any prior exposure to the systems that run our real-time scale—AWS, workflow orchestration (Airflow), containerization (Docker), or basic model deployment workflows—will accelerate your impact. What’s in It for You? At adjoe, you’re not here to just close JIRA tickets, you’re helping build the infrastructure behind one of the most impactful platforms in adtech. The systems you work on will reach hundreds of millions of users and power billions of decisions every day. - Go Big. Own projects with impact on 770M users and push adtech boundaries. - Move Fast. Ship solutions multiple times a day, learn from results, and keep momentum. - Be Direct. Solve problems openly and collaborate across teams. - Thrive Together. Grow with a diverse, global team of people from over 40 different countries that learn from each other. - Have Fun. Celebrate wins, enjoy daily victories, and bring your energy. Skip writing cover letters. Tell us about your most passionate personal project, your desired salary, and your earliest possible start date. We are looking forward to your application! We welcome applications from people who will contribute to the diversity of our Company. We welcome applications from people who will contribute to the diversity of our company.

▪️Разработка и развитие модуля Demand Planning. ▪️Проектирование и реализация высоконагруженных алгоритмов прогнозирования спроса и анализа рыночных сигналов.

Требования: — Опыт работы с Python не менее 5 лет, умение писать production-ready код и использовать паттерны проектирования. — Знание и опыт работы с Docker и Kubernetes. — Опыт работы с Apache Airflow. — Навыки работы с Apache Kafka и REST API. — Знание баз данных и распределённых вычислений. — Опыт использования GitLab CI или GitHub Actions. — Навыки мониторинга с Prometheus и Grafana. — Владение системой контроля версий Git. — Будет преимуществом понимание жизненного цикла MLOps. — Опыт работы с Feature Store и DVC. — Базовые знания машинного обучения, глубокого обучения и методов валидации моделей. — Опыт работы с Apache Spark и Hadoop. Локация: удалённо, Москва или Санкт-Петербург. Требуется гражданство РФ. Оформление через ИП или самозанятость.

Требования: — Необходим опыт работы с Python, SQL, PyTorch, Scikit-learn и Pandas. — Практические навыки использования FastAPI и Pydantic. — Опыт полного цикла разработки ML-моделей: от постановки задачи и подготовки данных до внедрения в production и поддержки. — Знание техник оценки ML-моделей, включая time-aware validation, backtesting, holdout datasets, calibration и error analysis. — Опыт создания CI-based regression suites для контроля качества моделей. — Навыки работы с model serving, MLOps, experiment tracking, model registries и контейнеризацией. — Опыт мониторинга моделей, выявления data drift и model degradation, а также организации переобучения. — Понимание вопросов data privacy, PII и тестирования на утечку приватных данных. — Опыт uncertainty quantification и работы с prediction intervals. — Дополнительным плюсом будут знания embeddings, retrieval systems, synthetic data generation, differential privacy, geospatial data, LangGraph, Airflow, Step Functions, а также опыт в сфере недвижимости и финансов. Задачи: — Разработка и внедрение ML-моделей для прогнозирования occupancy, headcount, rent и финансовых показателей. — Оценка качества моделей, проведение backtesting и анализ ошибок. — Разработка и поддержка ML-сервисов, API и MLOps-процессов. — Мониторинг production-моделей, контроль drift и degradation, организация переобучения. — Работа с embedding и retrieval-системами, synthetic data pipelines. — Взаимодействие с командами Product, Platform Engineering и Data Platform. Условия: — Только для фрилансеров, аустафф и рекрутинговые компании не рассматриваются. Требуемый уровень английского языка: B2. Опыт работы: от 5 лет на позиции Senior с навыками Python и SQL. География: LATAM.

Location: Москва, удалённая работа или гибрид, RU; Россия О команде Наша команда развивает направление оптимизации инференса LLM на GPU Авито. Мы повышаем эффективность использования оборудования и помогаем продуктовым командам запускать модели с учётом профиля нагрузки, качества и метрик производительности. Какие задачи вас ждут: проводить нагрузочное тестирование моделей по запросам продуктовых команд и оптимизировать их под конкретные сценарии использования; развивать методику оценки производительности LLM на нашем оборудовании: бенчмарки, инструменты и мониторинг; регулярно замерять производительность, отслеживать выход новых моделей и поддерживать таблицу результатов; тестировать новые GPU и сопровождать их настройку вместе с командами инфраструктуры и платформы инференса; масштабировать оптимизации на весь парк моделей: спекулятивное декодирование, мульти-LoRA, распределённый инференс, переход на новые инференс-бэкенды; доводить изменения до продакшена по циклу - профилировать, оптимизировать, тестировать, внедрять при подтверждённых метриках; запускать LLM-модели с использованием платформы инференса. Мы ждём, что вы: оптимизировали LLM-модели с учётом профиля нагрузки; работали с инференс-фреймворками LLM: vLLM, SGLang, TensorRT-LLM и другими, запускали и настраивали их, подбирали параметры; знаете, как устроен LLM-инференс, и понимаете методы оптимизации: prefill и decode, KV-кеш, batching, спекулятивное декодирование, квантизация; знаете принципы работы GPU: память, пропускная способность, утилизация, NVLink, конфигурации с несколькими GPU; профилировали GPU и работали с инструментами для мониторинга; проводили нагрузочное тестирование и бенчмаркинг LLM-моделей: latency, TTFT, TPOT, throughput, и тд, умеете корректно сравнивать результаты и делать выводы; работали с Docker, Kubernetes, метриками и дашбордами Prometheus, Grafana; готовы внедрять изменения в продакшен, поддерживать их и объяснять результат продуктовым командам. Будет здорово, если вы: писали или оптимизировали CUDA-кернелы, понимаете SIMT и архитектуру GPU NVIDIA; внедряли спекулятивное декодирование, мульти-LoRA-сервинг или распределённый инференс; работали с роутингом и балансировкой с учётом KV-кеша: llm-d, NVIDIA Dynamo, Mooncake или аналогами; знаете C++ или Go; следите за статьями и проектами с открытым исходным кодом по ускорению инференса и быстро проверяете идеи на практике. Работа у нас — это: расширенная забота о здоровье: ДМС, страхование жизни, онкострахование; сервис психологической поддержки, а также финансовые, юридические и ЗОЖ-консультации для вас и ваших близких; корпоративный спорт: групповые тренировки, велбиинг-приложение; спецпредложения от партнёров и скидки на сервисы Авито: бронирование жилья, услуги продвижения, ипотечное страхование; компенсация питания; инженерная свобода для реализации масштабных задач; продуктовый подход к инфраструктуре, который базируется на метриках; сильное инженерное комьюнити, готовое поддержать и поделиться опытом; эффективные инструменты для роста и развития: матрица компетенций, внутреннее обучение, менторство и обмен опытом.

Программа ML Residency — это возможность принять участие в исследованиях мирового уровня в сфере машинного обучения. Программа ML Residency — это возможность принять участие в исследованиях мирового уровня в сфере машинного обучения. Организаторы программы — Yandex Research и ВШЭ. Резиденты смогут заниматься прикладными и теоретическими исследованиями в области машинного обучения, работать над сложными проектами, публиковаться на NeurIPS, ICML, ICLR, CVPR и других ведущих научных конференциях, вносить вклад в высокотехнологичные сервисы. Участники программы будут сотрудничать с коллегами из Яндекса, ведущих мировых университетов и корпоративных лабораторий и смогут посещать курсы по машинному обучению в Школе анализа данных. Резиденты будут работать вместе с исследователями Yandex Research, которые выступят в роли наставников и кураторов. Совместно они выберут одну из проблем в области компьютерного зрения, речевых и диалоговых систем, обработки естественного языка или другой области машинного обучения, а затем разработают решение этой проблемы. Результатами работы могут стать научные публикации, внедрение в продукты Яндекса или проекты с открытым кодом. Чем вы будете заниматься: * Развивать техническое и теоретическое понимание современного машинного обучения * Изучать литературу по машинному обучению и погружаться в выбранную область * Учиться выявлять новые проблемы и предлагать их решения * Развивать навыки академического письма и устных выступлений Требования * Специалист с любым уровнем подготовки — от студента в области машинного обучения до опытного исследователя из смежных дисциплин, например: математики, физики, химии, компьютерных наук или лингвистики * Имеете хорошее фундаментальное образование, искренне увлечены машинным обучением и мечтаете изменить мир к лучшему

Location: Москва Мы создаем сервисы кредитования в Сбербанке, позволяющие клиентам быстро и удобно получать кредиты онлайн. Клиент может подать заявку на получение кредита через мобильное приложение или WEB, получить одобрение всего за две минуты и мгновенно перевести средства на свою банковскую карту или счет, без необходимости посещать отделение банка лично. Наш проект постоянно развивается. Мы планируем улучшать клиентский опыт и развивать наши сервисы внедряя механики динамического ценообразования. Работа в нашей команде позволит вам принять участие во всех этапах разработки высоконагруженных распределённых систем, начиная от проектирования и заканчивая внедрением в промышленную эксплуатацию. Вы сможете активно влиять на создание новых продуктов и улучшение существующих бизнес-процессов, применяя современные технологии и подходы. Возможность углублять знания и улучшать навыки в области современного программирования и управления проектами. Комфортные условия труда - современное оборудование и стабильность занятости в одной из крупнейших финансовых организаций страны. Присоединяйтесь к нашей команде и станьте частью инновационного процесса, создающего удобные решения для миллионов пользователей по всей России!  Первый этап отбора на эту вакансию — общение с AI-рекрутером. После отклика ждите сообщение от него, диалог займёт примерно 10 минут. Задача AI-рекрутера — уточнить недостающие детали и ускорить рассмотрение вашей кандидатуры. AI-рекрутер только начинает свой путь, поэтому просим относиться с пониманием. Ваш опыт и участие помогут сделать его удобным и полезным для всех! Обязанности * руководство командой DS (продуктовый и исследовательский треки) плюс собственная инженерная работа — от архитектуры ML/AI-решений до инструментов для fine-tuning LLM * управление командой DS: постановка и декомпозиция задач, приоритизация, контроль качества * менторство, код-ревью, стандарты разработки и исследований; участие в найме * архитектура ML/AI-решений, обоснование выбора методов и моделей * разработка инструментов fine-tuning LLM (LoRA/QLoRA): подготовка датасетов, пайплайны обучения, оценка качества * разработка и валидация ML-моделей, feature engineering, вывод в production * исследовательские гипотезы, дизайн экспериментов, статистическая оценка, прототипы новых подходов * предложения по решению бизнес-задач аналитическими методами, консультирование заказчика, презентация результатов. Требования * одинаково уверенная работа как исследователь/моделист и как инженер, доводящий AI-решения до работающего продукта * алгоритмы и структуры данных на уровне понимания внутреннего устройства, а не вызова библиотек: базовые (сортировки, поиск, деревья, графы, динамическое программирование, оценка сложности), классический ML (линейные модели, градиентный бустинг, SVM, кластеризация), DL (обратное распространение, оптимизаторы SGD/Adam, механизмы внимания, трансформеры), LLM-инженерия (токенизация BPE, LoRA/QLoRA, ANN-поиск) * умение разобрать алгоритм по статье и реализовать в коде * готовность самостоятельно разрабатывать скиллы и тулы для LoRA/QLoRA-адаптации LLM * Data Scientist / ML-инженер — от 4–5 лет * техническое лидерство или менторство — от 1–2 лет * вывод моделей в production и их сопровождение; исследовательская работа * Python (scikit-learn, pandas, numpy, PyTorch), SQL/PostgreSQL * классический ML, валидация моделей, статистика, математическое моделирование * NLP/LLM: трансформеры, Hugging Face Transformers, LLM API, LangChain; промпт-инжиниринг, RAG * практический опыт обучения нейронных сетей: трансформеры + минимум 2–3 типа архитектур (CNN, RNN/LSTM, VAE, RL-агенты) * MLOps: ML-пайплайны, Docker, FastAPI, Git, воспроизводимость экспериментов * готовность проводить код-ревью и давать развивающую обратную связь * способность понятно объяснять технические решения нетехнической аудитории * постановка задач команде, оценка трудоёмкости; готовность развивать управленческие навыки в процессе. Будет плюсом: * XAI, большие данные (PySpark, Hive, HDFS), банковская / финансовая сфера * опыт построения модели, давшей измеримый бизнес-эффект (метрики, выручка, экономия ресурсов) * участие в соревнованиях Kaggle или аналогичных. Условия * гибридный формат работы (4 дня офис) на Кутузовском, 32к1 * ежегодный пересмотр зарплаты, квартальная и годовая премия * более 400 образовательных программ СберУниверситета для профессионального и карьерного развития * программа адаптации и помощь руководителя на старте * расширенный ДМС, льготное страхование для семьи и корпоративная пенсионная программа * гибкий дисконт по ипотечному кредиту, равный 1/3 ключевой ставки ЦБ * подписка Прайм с возможностью совместного использования на трёх близких * вознаграждение за рекомендацию друзей в команду Сбера.

Location: Remote - Canada Narvar is Growing! We're looking for a leader to own the machine learning systems behind post-purchase commerce — identity, fraud, and the intelligence layer that agentic AI runs on. Hundreds of millions of consumers interact with Narvar every year, across 20B+ orders and 1,300+ retail brands. That data powers three systems: Graphite , our identity resolution engine that ties fragmented consumer data into a single verified identity across retailers; IRIS , our fraud and returns-abuse detection engine built on top of it. You'll own the ML organization behind all of it — the models, the platform they run on, and the people who build them. This is a role for someone who wants a system with real consequences: fraud decisions that move retailer margin, identity resolution that agents make automated decisions on, and an adversary on the other side who adapts every quarter. For this role, you should be located in Canada and able to work within EST/EDT OR PST hours. We are fully remote. Day-to-day Own the ML charter across identity resolution, fraud and abuse detection, risk scoring, and consumer intelligence — strategy, roadmap, and delivery Build and grow a high-performing, globally distributed team of ML engineers; hire, coach, and develop senior ICs and managers Set the technical bar for how models get built, evaluated, deployed, and monitored — and hold the org to it Push identity resolution coverage, precision, and profile classification accuracy against a measured, frozen-holdout baseline — not against vibes Own IRIS model performance end-to-end: detection rate, false-positive rate, label quality, and the feedback loops that keep both honest as fraud patterns shift Build the ML platform layer — feature stores, training pipelines, model registry, online serving, drift and performance monitoring — so model velocity isn't bottlenecked on infrastructure Partner with the AI engineering team so identity and risk signals are first-class inputs to NAVI's agent decisions Work directly with Product, Engineering, Security, and Customer Success to translate messy retailer problems into ML problems worth solving — and to say no to the ones that aren't Own build-vs-buy and data-partner decisions (third-party identity data, enrichment providers), including the economics Communicate model performance, risk, and tradeoffs credibly to executives, retailers, and the board What We're Looking For We care more about judgment and ownership than credentials. You're likely a strong fit if you: Have 12+ years in engineering with 5+ years managing ML or data teams, including managing managers or senior ICs Are an engineer at heart — you can still read a training pipeline, review a feature spec, and tell when an eval is measuring the wrong thing Have shipped ML systems that make consequential automated decisions in production, and have owned them after launch — drift, retraining, incidents, and all Have deep experience with at least one of: entity resolution / identity graphs , fraud and abuse detection , risk scoring , or anomaly detection at scale Understand what makes ML different from software: labels are noisy and delayed, systems fail silently, offline metrics lie, and last quarter's model is fighting last quarter's adversary Have opinions about evaluation — precision/recall tradeoffs on heavily imbalanced data, holdout hygiene, feedback loops where the model's own decisions contaminate future labels Have built or scaled ML infrastructure: feature pipelines, training orchestration, online model serving with real latency budgets, monitoring and alerting Are fluent in Python and comfortable in a modern data stack (Spark, Airflow or equivalent, streaming, cloud data warehouses — we run on GCP) Have hired well, retained well, and can point to engineers whose careers got better on your watch Excel in a fast-paced environment with a flat structure — you set direction by earning trust, not by mandate Have a BS/MS in computer science, statistics, or an equivalent background Bonus points These aren't hard requirements, but strong indicators: You've worked on identity resolution or graph systems against third-party consumer data (credit bureau, telecom, or similar), including the matching logic and confidence scoring underneath You've operated adversarial ML systems where attackers actively adapt to your detection You've built real-time inference paths where identity or risk has to be resolved inside a request budget You've worked in retail, payments, fintech, insurance, or marketplace trust & safety You've partnered closely with LLM/agent teams and understand how structured ML signals ground agentic decisions You've navigated privacy, consent, and data-governance constraints on consumer identity data You've done 0→1 — stood up a function, a platform, or a data partnership from nothing You use modern tooling (including AI-assisted development workflows) to increase leverage, not outsource thinking Why ML at Narvar? Because the data is genuinely rare and the problems are genuinely hard. The dataset is a moat. Cross-retailer visibility across 20B+ orders means we can see coordinated fraud patterns no single retailer can — a returns abuser hitting five brands looks like a good customer at each one. The problem is adversarial and unsolved. Returns fraud and policy abuse are growing faster than retail itself, and the counterparty adapts. There is no off-the-shelf answer. The stakes are measurable. Fraud loss, false-positive cost, identity coverage, resolution rate. Not vanity metrics. It feeds something bigger. As AI shopping agents enter commerce, every agent will need a trust layer for returns, claims, and delivery promises. Identity and risk are that layer — and you'd own it. The work is greenfield enough to matter. Real-time resolution, cold-start modeling, historical backfill, new data partnerships — these are decisions still being made, not legacy you inherit. Why Narvar (from an engineer's perspective)? Real scale, real customers, real consequences Startup-level ownership with platform-level impact Teams that value thinking, judgment, and responsibility Low ego, high trust, and room to do your best work We're on a mission to simplify the everyday lives of consumers. Post-purchase is a critical phase of the customer journey. That's why we created Narvar - a platform focused on driving customer loyalty through seamless post-purchase experiences that allow retailers to retain, engage, and delight customers. If you've ever bought something online, there's a good chance you've used our platform! From the hottest new direct-to-consumer companies to retail’s most renowned brands, Narvar works with GameStop, Neiman Marcus, Sonos, Nike, and 1300+ other brands. With hubs in San Francisco, Atlanta, London, and Bangalore, we've served over 125 million consumers worldwide across 10+ billion interactions, 38 countries, and 55 languages. Pioneering the post-purchase movement means navigating into the unknown. Our team thrives on this sense of adventure while nurturing a mindset of innovation. We're a home for big hearts and we leave our egos at the door. We work hard but we always make time to celebrate professional wins, baby showers, birthday parties, and everything in between. We are an equal-opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status. #LI-Remote Below is the estimated annual salary for this position and does not include the other components that make up a Narvar offer including: annual bonus, equity, and benefits. The range reflects the minimum and maximum target for new hire salaries for the position across Canada. Within the range, individual compensation packages are based on factors unique to each candidate, including but not limited to, skill set, education and certifications, and work location. Narvar Salary Range $240,000 — $260,000 CAD Please read our Privacy Policy to learn what personal information we collect in connection with your job application, and how we may use and share it.

Задачи: — разработка и развитие платформы анализа изображений на базе машинного обучения, автоматизирующей выявление дефектов и контроль качества при гальванической и финишной обработке поверхностей. Требования: — опыт работы от 5 лет в DevOps или Cloud с AWS в качестве основной облачной платформы; — практический опыт развертывания ML-моделей или Python-сервисов в AWS, включая SageMaker, inference endpoints и Step Functions; — знание английского языка на уровне Intermediate и выше.

Location: Remote Workplace: Remote We are building AI to simulate the world through merging art and science. We believe that world models are at the frontier of progress in artificial intelligence. Language models alone won’t solve the world’s hardest problems – robotics, disease, scientific discovery. Real progress requires models that experience the world and learn from their mistakes, the same way that humans do. And this kind of trial and error can be massively accelerated when done in simulation, rather than in the real world. World models offer the most clear path to general-purpose simulation, changing how stories are told, how scientific progress is made and how the next frontiers of humanity are reached. Our team consists of creative, open minded, caring and ambitious people who are determined to change the world. We aspire to continuously build impossible things and our ability to do so relies on building an incredible team. If you are driven to do the same, we'd love to hear from you. *Titles posted to our careers page are updated regularly to best advertise the role and expected seniority, but internally we have a flat titling system with two types of titles: one for individual contributors and one for people managers. ABOUT THE ROLE We're looking for an ML infrastructure engineer to own model evaluation at Runway, end to end. Every decision we make about a model – which checkpoint to keep training, what to ship to millions of users, which datamix and architecture shows the most promise – rests on evals. Today that work is spread across the organization. You'll turn it into one platform. Your job will be to design and build the systems that generate samples at scale, score them with automated metrics and human annotations, track results across checkpoints and releases, and put the answers in front of researchers in minutes rather than days. You'll define what "better" means operationally – how we measure it, how confident we are, and how a result becomes a ship/no-ship decision. You'll be embedded within research teams as a member of ML Platform, and you'll set the technical direction for evals across the company. This is an extremely high-leverage role: the quality of our models is bounded by how well we can measure them. A PEEK AT OUR TECHNICAL STACK Our inference and training code is written in Python (PyTorch) and is cloud-native. We leverage multiple clouds and use a combination of Kubernetes native and home-grown tooling for efficient job orchestration. We’re big users of agentic development and operations. We have a robust internal platform and provide broad access to the latest models and harnesses. You’ll have access to best-in-class models, agents, GPUs, storage and cloud services you need to drive a world class evaluation pipeline that helps us deliver frontier models. WHAT YOU'LL DO - Own the evaluation platform end to end: the tooling and systems required to generate, annotate, review and adapt at frontier scale - Define the CLIs, APIs, GUIs and storage layers required to make this process seamless, fast, sophisticated and collaborative - Work directly with research teams on video, image, audio, agents, and robotics to understand what they need to measure and build it, then generalize the result into the platform - Help set the standards for how we evaluate models at Runway: reproducibility, metric definitions, reporting formats, and when a result is trustworthy enough to act on - Support broad adoption of the platform and its integration throughout Runway’s research efforts across training, production model serving and more - Contribute broadly as a member of the ML Platform team to tools and systems that help Runway train and serve frontier models WHAT YOU'LL NEED - 5+ years of experience building ML infrastructure or data platforms in production environments, with at least some of that time spent on evaluation, experimentation, or benchmarking systems - Strong Python and PyTorch, and hands-on experience running large batch GPU workloads on Kubernetes - Experience designing data pipelines and storage for large volumes of media or model outputs, with attention to versioning and reproducibility - Familiarity with experimental statistics: paired comparisons, confidence intervals, multiple-comparison pitfalls, inter-rater agreement - Comfort building internal tools end to end, from the command line to the browser - Ability to lead a broad technical area: gather requirements from many teams, set direction, make tradeoffs, and drive a roadmap without waiting to be told - Familiarity with the full model development lifecycle: data, training, evaluation, serving - Self-starter who can work embedded with research teams and move fast - Strong systems thinking and pragmatic approach to production reliability - Humility and open mindedness; at Runway we love to learn from one another NICE TO HAVE - Track record of building evaluation suites for generative models - Hands-on work with LLM- or VLM-as-judge pipelines - Experience with online experimentation platforms and connecting offline metrics to product outcomes - Prior work evaluating agents or robotics policies Runway strives to recruit and retain exceptional talent from diverse backgrounds while ensuring pay equity for our team. Our salary ranges are based on competitive market rates for our size, stage and industry, and salary is just one part of the overall compensation package we provide. There are many factors that go into salary determinations, including relevant experience, skill level and qualifications assessed during the interview process, and maintaining internal equity with peers on the team. The range shared below is a general expectation for the function as posted, but we are also open to considering candidates who may be more or less experienced than outlined in the job description. In this case, we will communicate any updates in the expected salary range. Lastly, the provided range is the expected salary for candidates in the U.S. Outside of those regions, there may be a change in the range, which again, will be communicated to candidates. WORKING AT RUNWAY Great things come from great teams. https://www.youtube.com/watch?v=kwmj4ato2kw&ab_channel=Runway We’d love to hear from you. We’re committed to creating a space where our employees can bring their full selves to work and have equal opportunity to succeed. So regardless of race, gender identity or expression, sexual orientation, religion, origin, ability, age, veteran status, if joining this mission speaks to you, we encourage you to apply. More about Runway - Universal World Simulator https://runwayml.com/world-simulator.html - GWM-1 https://runwayml.com/research/introducing-runway-gwm-1 - Gen-4.5 https://runwayml.com/research/introducing-runway-gen-4.5 - General World Models https://runwayml.com/research/introducing-general-world-models - Robotics SDK https://runwayml.com/research/introducing-runway-gwm-1#robotics-section - Conversational Real-time Agents https://runwayml.com/research/introducing-runway-gwm-1#avatars-section - Runway Studios https://runwayml.com/studios We're excited to be recognized as a best place to work: Crain's https://www.crainsnewyork.com/awards/best-places-work-2023 | InHerSight https://www.inhersight.com/companies/best/city/new-york-city-ny | BuiltIn NYC https://builtin.com/awards/new-york-city/2024/best-places-to-work | INC https://www.inc.com/best-workplaces/2024

• Спроектировать и настроить AWS-архитектуру для ML-пайплайна: ingest, inference, хранение изображений, метаданных и результатов.

Location: Москва Workplace: Офис, Гибридный Яндекс Браузер помогает пользователям решать повседневные задачи. В работе мы объединяем LLM, классический ML и программные инструменты. Ищем ML‑разработчика, который будет превращать идеи в работающие решения: от подготовки данных и проверки гипотез до интеграции в Браузер и оценки качества. Мы создаём браузер, который помогает пользователям решать повседневные задачи: управлять вкладками, разбираться в содержимом страниц, заполнять формы и получать полезные подсказки в адресной строке. В работе мы объединяем LLM, классическое машинное обучение и программные инструменты. Ищем ML‑разработчика, которому предстоит превращать продуктовые идеи в работающие решения: от подготовки данных и проверки гипотез до интеграции в Браузер и оценки качества на реальных сценариях. Обязанности Улучшение автозаполнения форм Вы будете разрабатывать и оценивать способы определения назначения полей и подбора данных для заполнения, создавать воспроизводимую эмуляцию работы автофилла, выявлять причины ошибок и оценивать, насколько изменения улучшают работу. Автоматизация тестирования Браузера Вам предстоит разрабатывать агента, который будет воспроизводить пользовательские сценарии и проверять, исправлены ли ошибки. Также нужно создавать бенчмарки и автоматические оценки и проверять, насколько можно доверять выводам модели-судьи. Контекстные подсказки Алисы Вы будете классифицировать страницы и определять пользовательские задачи, чтобы в нужный момент предлагать полезные действия. Предстоит готовить разметку, строить модели и правила, оценивать точность и охват новых сценариев. Нужно будет исследовать и использовать перспективные инструменты классификации, например Jev, в сочетании с LLM. Развитие подсказок Вам предстоит улучшать классификацию запросов, ранжирование и смешивание подсказок из разных источников. Нужно готовить признаки и обучающие данные, проверять качество моделей и оценивать их влияние на пользовательский опыт. Развитие агента управления Браузером Вы будете учить агента понимать запросы пользователя и выполнять действия с вкладками, окнами и настройками. Вам предстоит подбирать модели, улучшать промпты и работу с инструментами, а также повышать надёжность сложных последовательностей действий. Больше об ML в Яндексе — в канале Yandex for ML Требования * Работали над ML‑проектами, которые дошли до продакшена * Знаете классические структуры данных и алгоритмы * Хорошо ориентируетесь в инструментах для решения NLP‑задач Будет плюсом * Строили системы на базе LLM * Активно используете агентов в повседневной жизни

Location: Singapore; Philippines; Albania; Ukraine; Indonesia; Australia; Serbia; Thailand; Khazakstan; Armenia; Türkiye; Romania Workplace: Remote Location: San Francisco or Singapore (on-site), or remote (Singapore or US Pacific time zones) Start Date: ASAP Languages: English Team: Engineering   ABOUT THE ROLE Our client is a fast-growing AI company building reliable, fair and open infrastructure for AI training data. Their marketplace helps the people who create data get value for it, and gives AI labs trustworthy data to train on. It is already used by startups, Fortune 500 companies and frontier labs, and the company is backed by leading venture investors, including Y Combinator. They are scaling quickly to meet strong demand. We are looking for a Research Engineer, Synthetic Data to build the synthetic data pipeline. You will turn domain-specific workflows into synthetic training tasks that are realistic enough to teach useful behaviour, structured enough to generate at scale, and difficult enough to expand model capabilities. This is a hands-on research engineering seat, not a commercial or privacy-systems role. You will design how synthetic tasks are created, mutated and validated, and you will analyse what those tasks actually teach models and agents. This is a full-time role, starting ASAP. Offices are in San Francisco and Singapore; remote candidates are welcome if their working hours can overlap about 70–80% with either San Francisco (US Pacific) or Singapore business hours. The exact daily schedule will be agreed during hiring.   WHAT YOU'LL DO - Work with subject-matter experts to create synthetic tasks for training AI agents across a range of professional and technical domains. - Design synthetic task-generation methods that produce diverse, realistic and learnable tasks. - Build systems and tooling to mutate, validate and improve synthetic tasks. - Analyse model and agent performance on synthetic tasks to understand what the tasks are teaching and where they fail. - Develop metrics to quantify and understand synthetic task diversity, realism, learnability and related qualities. - Take synthetic data pipelines from early research through to reliable systems, including in areas where the roadmap is not fully prescribed.   WHAT WE'RE LOOKING FOR - Proficiency in Python, Docker and Linux environments. - Experience with synthetic data research methods — please be ready to describe this in your application. - A strong understanding of what “good synthetic data” means, and of its limitations. - Experience building synthetic data pipelines end to end, including where there was no fully prescribed roadmap. - Experience working on environments, evals or benchmarks. - Clear written and spoken English, and comfort collaborating remotely across time zones. - Strong candidates may also be detail-oriented enough to catch subtle inconsistencies or edge cases in synthetic data; able to reason from first principles about task design, scoring and failure modes; and comfortable in unstructured problem spaces or early-stage environments. - Technical aptitude and learning potential matter more than years of experience. Motivated candidates are encouraged to apply even if they don't meet every point above.   WHY THIS ROLE MATTERS FOR YOU - Build the synthetic data pipeline: You will design the tasks and systems that generate training data at scale, not only run experiments on someone else's pipeline. - Research that has to work in production: Turn messy, domain-specific workflows into tasks that are realistic, measurable and usable for training agents. - See what the data actually teaches: Analyse model and agent performance, and improve the tasks when they fail. - High ownership, unstructured problems: You will build without a fully prescribed roadmap, from method design through tooling and metrics. - San Francisco, Singapore, or remote: Work on-site in San Francisco or Singapore, or remotely, with hours that overlap about 70–80% with either US Pacific or Singapore business hours. The exact daily schedule will be agreed during hiring.   Pragmatike is an Equal Opportunity Employer and is committed to providing equal employment opportunities to all applicants without discrimination. We recruit on behalf of our clients and prohibit discrimination and harassment based on race, color, religion, age, sex, national origin, disability status, genetics, protected veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by federal, state, or local laws. This policy applies to all terms and conditions of employment, including recruiting, hiring, placement, promotion, termination, layoff, recall, transfer, leaves of absence, compensation, and training. We are committed to a fair and inclusive hiring process. We process your personal data solely for recruitment purposes, in accordance with applicable privacy laws, and maintain reasonable safeguards to protect your information. Your data may be shared with our client(s) for hiring consideration, but will not be disclosed to third parties outside of the recruitment process.