Вакансии

Location: Almaty, Kazakhstan Workplace: hybrid Our AI Customer Support and Safety team is working on many areas including Computer Vision, NLP, RL etc, bringing business value to our products and improving the lives of our customers. Now we are aiming to strengthen several streams and looking for an experienced middle Machine Learning Engineer , who will help us in our journey. The primary focus for this position will be automation of the driver verification process, for which experience in computer vision is essential. Key Responsibilities Develop, deploy, and maintain machine learning models and systems Conduct in-depth data analysis, preparation, and processing Contribute to the design of ML systems, focusing on data-related aspects like annotation and processing Follow established security guidelines and identify potential risks Collaborate with peers to review and ensure designs are effective, flexible, and reusable Solve issues and provide feedback for engineering design and delivery processes Ensure that features are deployed according to requirements and operate without unintended side effects Create and maintain comprehensive documentation for models and processes Monitor and support deployed solutions to ensure they meet performance expectations Actively seek out information and internal solutions to ensure reusability and adoption of common technologies Our expectations 3+ years of experience in the Field. Previous software engineering experience is preferrable Strong proficiency in Python and its common data processing frameworks (e.g., for streaming, batch, and asynchronous data). Strong foundational knowledge in classical machine learning, deep learning, and advanced mathematics. Familiarity with MLOps principles and experience using MLOps tools for managing the ML model lifecycle. Experience with backend systems and infrastructure, and a willingness to learn new technologies (like Golang) for integration purposes. The ability to design solutions using common design patterns and various design tools. Experience in experiment design and the ability to validate the credibility of information from various sources. A solid understanding of the business value of delivered features. Strong problem-solving skills, with an emphasis on using technical facts and reasoning. Excellent communication and collaboration skills, including active listening and the ability to build consensus. A strong drive for continuous self-development and a collaborative mindset for contributing to communities. Ability to balance quality and speed depending on the case, and justify your decision Ability to explain the background and consequences of your technical decisions to non-technical stakeholders, including product and business managers English communication skills Why join us Help us challenge injustice by creating fair choices for millions of people across 47 countries. Develop your professional skills with access to mentoring, career consulting, and learning programs. Collaborate with teams around the world and gain international experience through our Global Talent Exchange Program. Engage in company-wide challenges, awards, sports activities, employee-led social impact and volunteering projects. Work alongside people who take initiative, speak openly, and challenge themselves to grow. Improve your language skills through co-financed courses and internal speaking clubs. Final benefits may vary depending on the location.

Location: Москва То, что многие компании считают своим прибыльным активом, зачастую является скрытым корнем финансовых проблем. Задача департамента по работе с проблемными активами – вовремя выявить трудности у клиентов и помочь преодолеть их. Используя передовые модели машинного обучения и искусственного интеллекта, мы создаем AI-профили клиентов, которые помогают сотрудникам быстрее разобраться в кейсе и выбрать наилучшую стратегию взаимодействия. В рамках нашего центра ИИ и комплексных решений вектор развития ориентирован на концептуальное переосмысление процессов урегулирования, автоматизацию рутинных процессов и создание дополнительного дохода в рамках работы с проблемной задолженностью. Обязанности * исследование клиентского опыта в рамках процессов оздоровления и взыскания, поиск «узких» мест в процессах и использование передовых инструментов для проведения локальных исследований (графовый анализ, LLM для поиска инсайтов, самовоспроизводящиеся ИИ-решения, disrupt-проекты) * регулярные встречи с Заказчиками для определения зон роста, предоставления непредвзятого взгляда на процессы и предложения к их трансформации * создание ИИ-решений, влияющих на клиентский опыт * построение чистого эксперимента и валидация ИИ-решения * проработка технологической реализации при поддержке команд разработчиков, data и OPS-специалистов * контроль внедрения ИИ-решения и последующего поведения решения в ПРОМ-контуре, анализ результатов внедрения * создание AI-агентов на базе LLM для оптимизации рутинных процессов: * речевая и текстовая аналитика взаимодействия с клиентом * выявление сущностей для автоматизации внутренних процессов * качественная суммаризация текста для составления профиля клиента * создание, обучение и внедрение моделей для оценки и прогноза финансового состояния клиента, скоринга, прогнозирования дефолта, антифрода и выявления мошенничества. Разработка систем подбора индивидуальных условий для урегулирования задолженности, а также применение What-If / Uplift моделирования и каскадов моделей для выстраивания процессов с использованием ИИ * графовый анализ компаний, их руководителей и бенефициаров, анализ новостного контента и других источников информации о клиентах. Требования * опыт работы 3+ лет в традиционных направлениях машинного обучения (ML), обработки естественного языка (NLP), глубокого обучения (DL) * знание библиотек и фреймворков: pandas, pyspark, catboost, pytorch, tensorflow, langchain/langgraph * опыт валидации AI-решений и их "боевого" пилотирования решений * прикладной опыт в использовании LLM, контроле качества ответов и выстраивания графовой архитектуры с использованием LLM tools / MCP * знания математической статистики и основ A/B-тестирования. _Будет плюсом:_   * опыт руководства проектами или самостоятельного обеспечения всего цикла производства ИИ-решения от гипотезы до реализации * знание основ финансового анализа и бухгалтерского учета * опыт курирования или наставничества, работа в крупных AI-проектах с смежными командами * английский язык — уровень Intermediate и выше * опыт в проектах, связанных с клиентским опытом / бизнес-анализом / исследованием очень больших данных * опыт использования AI-инструментов в разработке (LLM-ассистенты для написания и ревью кода, агенты для автоматизации рутинных операций, инструменты автодополнения вроде GitHub Copilot). Условия * возможен гибридный формат работы, локация офиса: Москва, Кутузовский пр., д.32, к.3 * ежегодный пересмотр зарплаты, годовая премия * корпоративный спортзал и зоны отдыха * более 400 образовательных программ СберУниверситета для профессионального и карьерного развития * 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: Москва Команда занимается разработкой «AI-помощника» - это проект с использованием LLM, который призван радикально улучшить клиентский опыт при при оформлении основных продуктов банка и помочь клиенту разобраться в продуктах. Обязанности * проведение исследовательского анализа данных (EDA) для выявления ключевых метрик и закономерностей, построение моделей. Data-driven подход к разработке * разработка и внедрение AI-агентов на базе LLM (SOTA open-source решения), в т.ч участие в создании и развитии RAG-систем (Retrieval-Augmented Generation) * работа с LLM: написание и оптимизация промптов (prompt engineering), управление взаимодействием с моделью * интеграция решений с корпоративными системами, обеспечение их взаимодействия * создание автономных бизнес-процессов в компании. Требования * опыт работы с Python и библиотеками: pytorch, pandas, numpy, SQLAlchemy, Scikit-learn, Matplotlib/Seaborn * писать поддерживаемый, читаемый код; участвовать в ревью, обсуждениях и формировании технических подходов * практический опыт в NLP: векторизация текста, классификация запросов, построение RAG-систем, дообучение языковых моделей, эмбеддингов и пр. * понимание классических алгоритмов ML * онимание методов обучения нейросетевых моделей (transfer learning, fine tuning) * навыки работы с Git, опыт интеграции решений в промышленные среды. Условия * офисный формат работы - офис на Кутузовской, сзонами отдыха и спортзалом; * годовой бонус и ежегодный пересмотр * социальный пакет (ДМС, страхование от несчастных случаев, льготное кредитование, подключение родственников к полису) * корпоративный университет Сбера, внутренняя образовательная платформа, участие в IT-конференциях * льготная ипотека в Сбере, корпоративная пенсионная программа, подписка СберПрайм+ с возможностью использования до трех близких, скидки от партнеров * материальная помощь при наступлении значимых событий в жизни; * дисконт-программы от компаний партнеров (фитнес, страхование, туризм) * подарки и мероприятия для детей сотрудников. Оздоровительные программы для детей сотрудников.

Location: Москва Дивизион "Моделирование и исследование данных" отвечает за компетенцию Data Science в Блоке. Мы разрабатываем и внедряем end-to-end решения на основе машинного обучения в бизнес-процессы Сети продаж банка. Сейчас мы ищем в команду MLE Data Scientist’а. Обязанности * Ведение всех этапов AI-проекта: от постановки задачи и требований с бизнес-заказчиком до внедрения и мониторинга; * Разрабатывать с нуля AI-агентов и микросервисы для AI-продуктов; * Разрабатывать алгоритмы взаимодействия ML-моделей и сервисных модулей; * Принимать участие в разработке архитектуры и дизайна агентов, ML приложений и AI-сервисов; * Проектировать и разрабатывать backend-части наших систем; * Стэк: 50% Classic ML + 50% Agents. Требования * Компетенции: Python, DE, ML (Classic ML + NLP), Statistics, LLM/Agents, MLOps; * Развитые soft-skill для общения с заказчиком; * Понимание принципов разработки, внедрения и сопровождения AI-решений; * Уметь писать хороший и понятный пром-ready код; * Опыт промышленной разработки на python от 2-ух лет; * Хорошей знание алгоритмов, структур данных, GIL, async; * Опыт работы с реляционными и нереляционными базами данных, понимание границ применимости каждого вида из них; * Опыт построения отказоустойчивых высоконагруженных распределенных систем; * Будем иметь дело с широким набором инструментов и технологий: Python, Elastic, Kafka, FastAPI, Docker, OpenShift, SQL, Spark, Hadoop, GPU, BitBucket. Условия * Полное соблюдение ТК РФ, льготные условия по программам страхования, кредитования для сотрудников Банка;) * График 5/2, офис на м. Кутузовская, гибрид; * Достойный уровень вознаграждения (оклад + внушительная годовая премия); * ДМС с первого дня работы и льготная мед. страховка для близких родственников; * Социальная поддержка сотрудников; * Возможность стать частью команды, реализующей программы цифровой трансформации банка - на основе передового опыта и инновационных идей; * Широкие возможности для профессионального развития: корпоративный университет и множество онлайн-программ обучения. Корпоративное обучение за счет компании; * Бесплатная подписка СберПрайм с возможностью подключения 3 родственников, многочисленные скидки и бонусы от партнеров: СберМаркет, МегаМаркет, Самокат, Еаптека и др.; * Корпоративная пенсионная программа

Location: California Workplace: onsite Overview We’re hiring data collection operators to support physical AI programs. In this role, you’ll operate physical AI data collection systems (via teleoperation tools or guided interfaces) to generate high-quality real-world data used to train and evaluate AI models. This is hands-on, execution-focused work. You’ll be supporting a robotics data collection pilot using I2RT YAM bimanual robotic arms. Each station is staffed by two people, and you’ll carry out assigned teleoperation and human-robot interaction tasks to help capture reliable, consistent data.  Project Details Location: Palo Alto, CA — 100% onsite Type: Full-time, short-term W-2 employee Schedule: Full-time; 8 hours/day, 5 days/week. Must be able to stand for extended periods during shifts  Duration: Minimum 2-week engagement; potential extension through year-end and beyond (extension not guaranteed)  Start: ASAP Compensation: US$35/hour Work authorization: Must be authorized to work onsite in the U.S. Key responsibilities Operate physical AI data devices, including teleoperating bimanual robotic arms (I2RT YAM) to execute scripted tasks precisely, following SOPs. Perform human robot interaction tasks per script, including gestures and object handovers. Follow detailed instructions and documented SOPs, adapt to evolving guidelines, and complete hands-on training, applying it consistently. Coordinate closely with your station partner to maintain timing, consistency, and safety. Capture and validate data. Record trial status, notes, and anomalies as directed; promptly flag issues and assist with resets/retries. Adhere to onsite safety, privacy, and data handling protocols; provide constructive feedback to improve procedures. Support fast iteration cycles as tasks, tools, and priorities evolve. Qualifications Bachelor’s degree or equivalent work experience Experience with gaming, simulation tools, or technical systems is a plus Strong hand-eye coordination and spatial awareness Detail oriented, reliable, and consistent in repetitive, process-driven work. Clear communication and teamwork; receptive to coaching and feedback. Prior teleoperation/robotics or lab data collection experience preferred; training provided. Basic computer skills for simple logging and spreadsheets. Preferred Qualifications Exposure to robotics, autonomous systems, drones, or simulation environments Experience with teleoperation, remote control systems, or QA/testing workflows Familiarity with data collection, logging, or annotation processes Ability to quickly learn new software applications Interest in AI, robotics, or emerging technologies Experience working in structured, metrics-driven environments Please note that in order to verify work authorization as is required by Federal law (I-9 process), all new employees must complete a live video verification with their selected IDs and provide photos of these selected IDs within their first 3 days of employment.   To know more details (Click here)   In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification form upon hire.  In addition, we employ anti-fraud checks to ensure all candidates meet the requirements of the program.     As a trusted global transformation partner, Welocalize accelerates the global business journey by enabling brands and companies to reach, engage, and grow international audiences. Welocalize delivers multilingual content transformation services in translation, localization, and adaptation for over 250 languages with a growing network of over 400,000 in-country linguistic resources. Driving innovation in language services, Welocalize delivers high-quality training data transformation solutions for NLP-enabled machine learning by blending technology and human intelligence to collect, annotate, and evaluate all content types. Our team works across locations in North America, Europe, and Asia serving our global clients in the markets that matter to them. www.welocalize.com   To perform this job successfully, an individual must be able to perform each essential duty satisfactorily. The requirements listed below are representative of the knowledge, skill, and/or ability required. Reasonable accommodations may be made to enable individuals with disabilities to perform essential functions.

Location: California Workplace: onsite Overview We’re hiring an onsite lead to run day-to-day execution for our robotics data collection pilot, built around I2RT YAM bimanual robotic arms. You’ll guide a team of data collection operators, keep operations safe and consistent, and act as the main onsite point of contact for client stakeholders. Please note that this role does not own staffing schedules or inventory management.   Project Details Location: Palo Alto, CA — 100% onsite Type: Contract Schedule: Full-time; 8 hours/day, 5 days/week Duration: Minimum 2-week pilot; potential extension through year-end and beyond Start: ASAP Compensation: $50/hour Work authorization: Must be authorized to work onsite in the U.S. Key responsibilities Oversee daily execution and station readiness; enforce SOPs, quality, and safety protocols. Coach operators on teleoperation and HRI execution; provide on-the-spot refreshers and corrective guidance. Monitor live sessions and data quality; perform first-level troubleshooting and escalate issues appropriately. Maintain concise daily notes and issue trackers; communicate blockers, risks, and needs to client teams. Contribute to SOP and checklist refinements based on observations. Utilize client-provided equipment, SOPs, examples, and hands-on training to align the team on best practices. Qualifications Hands-on experience leading robotic teleoperation and/or HRI data collection. Strong communication, coaching, and problem-solving skills in lab/field environments. Comfortable driving consistency, safety, and quality in repetitive workflows. Please note that in order to verify work authorization as is required by Federal law (I-9 process), all new employees must complete a live video verification with their selected IDs and provide photos of these selected IDs within their first 3 days of employment.   To know more details (Click here)   In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification form upon hire.  In addition, we employ anti-fraud checks to ensure all candidates meet the requirements of the program.     As a trusted global transformation partner, Welocalize accelerates the global business journey by enabling brands and companies to reach, engage, and grow international audiences. Welocalize delivers multilingual content transformation services in translation, localization, and adaptation for over 250 languages with a growing network of over 400,000 in-country linguistic resources. Driving innovation in language services, Welocalize delivers high-quality training data transformation solutions for NLP-enabled machine learning by blending technology and human intelligence to collect, annotate, and evaluate all content types. Our team works across locations in North America, Europe, and Asia serving our global clients in the markets that matter to them. www.welocalize.com   To perform this job successfully, an individual must be able to perform each essential duty satisfactorily. The requirements listed below are representative of the knowledge, skill, and/or ability required. Reasonable accommodations may be made to enable individuals with disabilities to perform essential functions.

Развитие скоринговых моделей. Построение предиктивных моделей для задач B2C и B2B кредитования, страхования и e-commerce.

• Ручное тестирование и валидация изменений кода, сгенерированных агентной/автоматизированной системой на основе транскриптов поддержки (чат/голос). • Проведение exploratory testing и проектирование подробных тест‑сценариев для сложных изменений.

Location: Remote, USA Workplace: remote At PointClickCare our mission is simple: to help providers deliver exceptional care. And that starts with our people. As a leading health tech company that’s founder-led and privately held, we empower our employees to push boundaries, innovate, and shape the future of healthcare. With the largest long-term and post-acute care dataset and a Marketplace of 400+ integrated partners, our platform serves over 30,000 provider organizations, making a real difference in millions of lives. We also reinvest a significant percentage of our revenue back into research and development, ensuring our employees have the resources to innovate and make a lasting impact. Recognized by Forbes as a top private cloud company and honored as one of Canada’s Most Admired Corporate Cultures, we offer flexibility, growth opportunities, and meaningful work.  At PointClickCare, we empower our people to be the architects of a smarter healthcare future; one that is human-first and accelerated by AI to create meaningful and lasting change. Employees harness AI as a catalyst for creativity, productivity, and thoughtful decision-making. By integrating AI tools into our daily workflows, collaboration is enhanced, outcomes are improved, and every team member has the proficiency to maximize their impact. It all starts with our hiring practices where we uncover AI expertise that complements our mission, and we continue to invest in training and development to nurture innovation throughout the employee journey. Join us in redefining healthcare — so it doesn’t just survive, it thrives. To learn more about PointClickCare, check out Life at PointClickCare and connect with us on Glassdoor and LinkedIn. This is a part-time contract opportunity that may serve as a supplemental engagement for individuals currently in full-time roles.    The position offers flexible scheduling, with a commitment of up to 28 hours per week. An average of 8-10 hours per week   3-month contract   - The role is largely (95%) completing labeling of PointClickCare clinical notes for data input to help train AI models. There are often singular project tasks at hand that requires sitting for extended periods of time while labeling data. - While labeling, provide clinical feedback and recommendations using nursing knowledge of Senior Living/SNF electronic medical record - Provide Clinical expertise within the labeling team. Ability to align with set labeling rules as well as provide challenging input regarding potential labeling changes with the goal of improving the product. - Work with a team of subject matter experts involving quality review of AI products. -Complete the required training as defined by security, clinical director and HR. Requirements Clinical background ( RN, NP) – at least 5 years' acute or LTPAC clinical experience, clinical software experience and predictive intelligence with an EHR preferred A general familiarity with how AI is being used in healthcare and understanding terminology such as LLM, NLP, hallucinations Clinical degree required; bachelor's degree preferred with proof of clinical license verification Strong organizational and project management skills Must be a self-starter and work independently Must be a multitasker with excellent communication skills who is energized by working in a complex, dynamic, and fast-paced environment Experience in data labeling, supporting and servicing customers using a clinical software application is preferred Technical acumen or familiarity with the PointClickCare EHR is preferred but not required Passion for transforming healthcare Must have a working computer and internet and be willing to comply with technical requirements as outlined by our security and technical teams Specifics Include: Physical security requirements while working within the job, including a private space, a properly secured Wi-Fi connection, no shared access to the computer used for labeling Install the approved and assigned VPN and use a particular internet browser as defined by the project Possess strong proficiency in written and spoken English with professional communication skills and the ability to understand medical terminology Must be able to complete the required background check, contract period and follow HR policies while representing PCC and the period that follows based on policy. #LI-AV1 #LI-remote It is the policy of PointClickCare to ensure equal employment opportunity without discrimination or harassment on the basis of race, religion, national origin, status, age, sex, sexual orientation, gender identity or expression, marital or domestic/civil partnership status, disability, veteran status, genetic information, or any other basis protected by law. PointClickCare welcomes and encourages applications from people with disabilities. Accommodations are available upon request for candidates taking part in all aspects of the selection process. Please contact recruitment@pointclickcare.com should you require any accommodations. As part of our commitment to a streamlined and equitable hiring experience, PointClickCare uses AI tools to assist with candidate screening and assessment. When you apply for a position, your information is processed and stored with Lever, in accordance with Lever’s Privacy Policy. We use this information to evaluate your candidacy for the posted position. We also store this information, and may use it in relation to future positions to which you apply, or which we believe may be relevant to you given your background. When we have no ongoing legitimate business need to process your information, we will either delete or anonymize it.  If you have any questions about how PointClickCare uses or processes your information, or if you would like to ask to access, correct, or delete your information, please contact PointClickCare’s human resources team: recruitment@pointclickcare.com  PointClickCare is committed to Information Security. By applying to this position, if hired, you commit to following our information security policies and procedures and making every effort to secure confidential and/or sensitive information.

Location: Remote Workplace: Remote STAFF PRODUCT MANAGER, PRODUCT DISCOVERY ENGINE ABOUT US Constructor powers product search and discovery for some of the largest retailers in the world. We serve billions of requests every week, and you’ve probably seen our results somewhere and used our product without knowing it. We differentiate ourselves by focusing on metrics over features, and reinventing search and discovery from the ground up as a machine learning challenge with the specific goal of improving metrics like revenue. We’re approximately doubling year over year despite the market slowing down and have customers in every eCommerce vertical. We’re a passionate team of technologists who love solving problems and want to make our customers’ and coworkers’ lives better. We value empathy, openness, curiosity, continuous improvement, and are excited by metrics that matter. We believe that empowering everyone in a company to do what they think is best can lead to great things. JOB SUMMARY We’re looking for a systems-minded Staff Product Manager to take Constructor’s search and discovery to the state of the art — and keep it there — across every retail industry we serve. This is a highly technical, high-leverage role at the core of our Product Discovery Engine chapter: the teams responsible for how we retrieve, understand, and rank the right products for every shopper, on every query, for every customer. Your job is to raise discovery quality and, through it, our customers’ business metrics. You won’t be measured on features shipped — you’ll be measured on relevance and business outcomes, like conversion, revenue per visit, and search-attributed GMV, moving up. You’ll start embedded with our Search Quality team, owning its strategy and roadmap and getting deep into the systems that decide what shoppers actually see. Over time, your mandate grows to setting direction for the entire Product Discovery Engine chapter — deciding where we’re heading and why across recall, ranking, query understanding, personalization, and the ML infrastructure underneath. This is a Staff role today with a clear path to Director as you take on chapter-wide leadership. Search at Constructor is fundamentally a machine learning problem, so this role lives and breathes ML. You’ll partner closely with ML, data science, and engineering to turn the latest research into production wins, keep us ahead of both commercial competitors and academic state of the art, and be the connective tissue that aligns multiple teams — ML Recall, ML Infra, and, over time, teams like CORE, SABR, and Catalog Management — behind a single, coherent vision for great discovery. If you’re excited about driving innovation across dozens of the world’s largest retailers, being held to relevance and business outcomes, and pushing an already best-in-class product further ahead, this role is for you. WHAT YOU’LL DO - Own business impact, not features. Drive measurable lift in search-attributed conversion, revenue per visit, and GMV alongside relevance across our customer base. - Deliver across every retail industry. Move quality and business metrics across all the verticals we serve, accounting for the differences in catalog, language, and shopper intent. - Keep us at the state of the art. Continuously translate the latest ML / IR research and competitive moves into production wins, and ship SOTA ML / LLM-assisted capabilities that beat our previous best on relevance. - Set chapter vision and strategy. Define the multi-quarter direction for the Product Discovery Engine — where we’re heading and why — tie it directly to customer business outcomes, and rally teams behind it. - Lead Search Quality first. Own its strategy and roadmap, driving measurable improvements in the relevance and correctness of results. - Orchestrate across the pipeline. Unify recall, ranking, query understanding, and personalization behind one definition of quality and one set of business metrics, so improvements compound instead of colliding. - Turn ML research into production wins. Partner deeply with ML and data science to take models — from candidate generation and recall to LLM-assisted interpretation — from idea to production, balancing accuracy, latency, and cost. - Make quality measurable and explainable. Build the metrics, evaluation frameworks, and tooling the whole chapter uses to measure, attribute, and explain discovery quality and its business impact. - Own quality under pressure. Lead investigation and triage when issues arise, and design durable, systemic fixes rather than one-off patches. - Grow the mandate. Expand into chapter-wide leadership spanning ML Recall, ML Infra, and — over time — teams like CORE, SABR, and Catalog Management. WHAT SUCCESS LOOKS LIKE Within your first year, you will have: - Driven measurable lift in search-attributed conversion and revenue per visit — moving business metrics, not just quality metrics. - Driven measurable business-metric lift (conversion, revenue per visit, GMV) and relevance lift across every major retail vertical we serve. - Shipped at least one state-of-the-art ML / LLM-assisted capability into production that beats our previous best on relevance. - Won head-to-head quality benchmarks against every major competitor in the verticals we compete in, with a repeatable, defensible methodology. - Taken core quality metrics to best-in-class levels and kept them improving quarter over quarter. - Stood up an evaluation framework and tooling that ties model and system investments directly to customer business outcomes. WHAT WE’RE LOOKING FOR - Extensive product management experience (typically 8+ years) in search, discovery, ML, or NLP-heavy products, including time operating at a senior or staff level. - A track record of moving business metrics at scale — conversion, revenue, engagement — not just shipping features, and of setting strategy across multiple teams. - Deep understanding of the end-to-end search pipeline: query interpretation → recall / candidate generation → ranking → reranking → personalization → delivery. - Strong ML product depth — owning production models and reasoning about tradeoffs between high-capacity (LLM / batch) and low-latency (real-time) approaches — and a habit of staying current with state-of-the-art research. - Fluency with retrieval and recall, query understanding (tokenization, synonyms, spelling correction, multilingual support), and search evaluation frameworks. - A rigorous, measurement-first mindset that ties every quality investment back to business outcomes. - Excellent cross-functional leadership — able to align teams with different goals behind a shared definition of quality and a shared set of business metrics, working shoulder-to-shoulder with engineering, data science, and ML Ops. - Comfort operating with ambiguity, setting direction where none exists yet, and leading through clarity and credibility rather than the org chart. BENEFITS - Unlimited vacation time — we strongly encourage all of our employees to take at least 3 weeks per year. - A competitive compensation package including stock options. - Company-sponsored US health coverage (100% paid for employee). - Fully remote team — choose where you live. - Work-from-home stipend! We want you to have the resources you need to set up your home office. - Apple laptops provided for new employees. - Training and development budget for every employee, refreshed each year. - Parental leave for qualified employees. - Work with smart people who will help you grow and make a meaningful impact. This position is fully remote — Constructor.io http://Constructor.io is a remote-first company. DIVERSITY, EQUITY, AND INCLUSION AT CONSTRUCTOR At Constructor.io http://Constructor.io we are committed to cultivating a work environment that is diverse, equitable, and inclusive. As an equal opportunity employer, we welcome individuals of all backgrounds and provide equal opportunities to all applicants regardless of their education, diversity of opinion, race, color, religion, gender, gender expression, sexual orientation, national origin, genetics, disability, age, veteran status or affiliation in any other protected group. How we handle your data. By submitting this application, you acknowledge that Constructor will collect and process the personal data you provide for the purpose of evaluating your candidacy for this role. If you are not selected, we will retain your data for a limited period as described in our Recruitment Privacy Notice https://constructor.com/recruitment-privacy-notice. After that period your data will be deleted or anonymized.

Знание Python и основ машинного и глубокого обучения (ML/DL). Понимание NLP и LLM, принципов дообучения моделей (Fine-tuning), построения RAG-систем и семантического поиска.

ML-команда развивает рекомендательную систему и инструменты компьютерного зрения, обработки текста и распознавания речи. Предстоит улучшать модели, участвовать в сборе данных и оптимизировать инфраструктуру запуска моделей и мониторинга.

Location: Москва Команда «Мотивационный профиль» создает новый подход к работе с клиентом, который позволит перейти от классической персонализации к более глубокому пониманию человека на основе психологии. Мы извлекаем из данных психометрические характеристики — ценности, страхи, особенности характера и принятия решений — и используем их в продуктах банка и экосистемы. Для построения мотивационного профиля мы используем анкеты, подробные интервью, поведенческие и транзакционные данные. Работаем с известными психометрическими методиками и подходами, разработанными исследовательской командой. Одна из ключевых задач — научиться надежно определять психометрические характеристики на ограниченных выборках, масштабировать их на широкую клиентскую базу и использовать в рекомендательных сценариях и моделировании клиентских реакций. Обязанности Моделирование психометрических признаков: * подбирать постановку задачи в зависимости от объема и качества данных: supervised learning, weak supervision, few-shot и другие подходы * обучать модели для предсказания психометрических и поведенческих характеристик клиента * оценивать модели как по стандартным ML-метрикам, так и по тому, насколько хорошо их результаты работают в прикладных сценариях — сегментации, персонализации и моделировании поведения клиента Работа с текстовыми и поведенческими данными * строить пайплайны обработки текста: чанкинг, построение эмбеддингов, кластеризация и извлечение признаков * формировать датасеты для обучения, валидации и тестирования моделей * искать цифровые признаки, которые могут помочь определить характеристики мотивационного профиля. Требования * опыт работы Data Scientist / ML Engineer от 2–3 лет * уверенное знание Python и стандартного DS/ML-стека: pandas, numpy, scikit-learn, CatBoost / LightGBM / XGBoost * опыт построения supervised ML-моделей для табличных данных * опыт работы с текстовыми данными и NLP: embeddings, sentence-transformers, кластеризация, классификация текстов, семантический поиск * понимание основных подходов к работе с LLM: prompting, few-shot, zero-shot, LLM-разметка и оценка качества LLM-ответов. Условия * комфортный современный офис рядом с м. Кутузовская * офисный формат работы * ежегодный пересмотр зарплаты, годовая премия * корпоративный спортзал и зоны отдыха * 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: Москва Команда DataPlatform-Tools реализует платформенные инструменты для сервисов, AI-агентов и потребителей в домене HR-аналитики всего Банка. Мы обеспечиваем self-service для инженеров, аналитиков данных, DS, ML-инженеров: готовые DAG пайплайны, тесты качества данных (DQ), lineage, BI-визуализации. Разработка интеллектуальных ассистентов и чат-ботов для автоматизации процессов подбора персонала и карьерного развития сотрудников. Наш фокус — доступность данных не только для сотрудников, но и для AI-агентов! Для этого мы применяем подходы топовых компаний в мире и формируем собственную практику на огромном массиве данных Обязанности * развитие текущих LLM-агентов и решений на их основе * проектирование и разработка новых LLM-агентов и интеллектуальных ассистентов * работа с RAG-системами и векторными базами данных * оценка и повышение качества ML-моделей * участие в создании чат-ботов для автоматизации HR-процессов Требования * опыт работы с LLM и фреймворками для разработки агентов (LangChain, LangGraph и др.) * уверенные знания Python и SQL * опыт работы с NLP-задачами * опыт создания чат-ботов и интеллектуальных ассистентов * понимание принципов RAG (Retrieval-Augmented Generation) * опыт работы с векторными базами данных * навыки оценки качества ML-моделей Условия * комфортный современный офис рядом с м.Кутузовская * возможность выбрать удобный график – офис/гибрид * ежегодный пересмотр зарплаты, годовая премия * корпоративный спортзал и зоны отдыха * более 400 образовательных программ СберУниверситета для профессионального и карьерного развития * программа адаптации и помощь руководителя на старте (для Junior позиций) * расширенный ДМС, льготное страхование для семьи и корпоративная пенсионная программа * льготная ипотека для каждого сотрудника * бесплатная подписка Прайм с возможностью совместного использования на трёх близких * вознаграждение за рекомендацию друзей в команду Сбера. * корпоративная пенсионная программа * отсутствие строгого дресс-кода.

Твои задачи — RAG по методике и видеоурокам, файн-тюнинг LLM на русском и казахском, персонализация под профиль ребёнка. • Свой опыт файн-тюнинга (LoRA / QLoRA): сам собрал датасет, дообучил модель, измерил результат.

-Участвовать в разработке AI-решения для анализа документов, изображений и текста (KYC-верификация), включая работу с Computer Vision (CV), OCR и NLP. -Подбирать и адаптировать модели под бизнес-задачи.

Location: Limassol, Cyprus Workplace: fully Sumsub is the first AI-powered trust infrastructure for compliance operations at scale. It connects identity and business verification, fraud prevention, transaction monitoring, and risk workflows, helping teams reduce manual work and enter new markets. Trusted by over 4,000 clients across financial services, crypto, mobility, trading, marketplaces, education, and iGaming, including Bybit, Bitpanda, Wirex, Avis, Vodafone, Duolingo, and Kaizen Gaming. Sumsub is certified as a Great Place To Work® in the US, recognized as a TIME & Statista Excellent Workplace, honored with the TrustRadius Tech Cares Award for Diversity & Inclusion, and rated 4.7 out of 5 on Glassdoor. Our 1,000+ people around the world back it up, too: 92% are proud to be Sumsubers, and 90% say their work is interesting and challenging. Now we are looking for a AI Engineer (LLM focus) to build and scale the core intelligence behind Summy AI Copilot , our agentic assistant for compliance officers. This is an exciting opportunity to own a cutting-edge GenAI product, pushing LLMs beyond basic text generation to execute complex reasoning, and autonomous tasks. What You Will Be Doing: Drive Summy AI Copilot: Own the core intelligence behind our AI assistant for compliance officers, taking features from concept to production rollout and operation. Agentic Orchestration: Design and build robust, multi-turn AI agents (using frameworks like LangGraph) capable of complex reasoning, context retention, and autonomous action. LLM Observability & Evaluation: Set up comprehensive tracing, monitoring (e.g., LangFuse), and evaluation pipelines (LLM-as-a-judge) to track quality, latency, and token cost. Agentic Experience: Evolve our public MCP server and Agent Skills so that external AI agents (Claude, ChatGPT, etc) can work with Sumsub reliably and securely. Pragmatic Iteration: Start with elementary, solid baselines and iteratively scale complexity based on real-world production metrics and A/B tests. About You: Experience: 5+ years in Software/Data/ML Engineering, with 1+ years explicitly focused on building and shipping production-grade GenAI/LLM systems. In-Depth LLM Knowledge: You understand how LLMs work under the hood. You know how to push models far beyond basic text generation (e.g., structured extraction, tool calling, cost-aware routing). LLM Stack Mastery: Hands-on experience with agentic orchestration (LangChain/LangGraph) and modern LLM APIs (Claude, Gemini, OpenAI). Engineering Excellence: You write production-ready code for microservices (Java/Dropwizard), with experience in cloud deployments (GCP/AWS, Docker, Kubernetes). Ownership: You own the system from early talks with customers to operating it in production. Nice to have Background in classical NLP, traditional MLOps, or large-scale data engineering. Familiarity with compliance, anti-fraud, KYC/AML, or fintech domains. Experience shipping production RAG systems. What We Offer: Remote-first, trust-based culture. Work from the place that works best for you. No mandatory office days, no attendance trackers. In some locations, we provide offices or coworking spaces, but the choice is yours. True flexibility. We do not fix you to a 9-to-5 schedule. You can adjust your working hours when needed, as long as your day stays productive and in sync with the team. Extra time off. Your birthday is a holiday here. Add to that 10 personal days each year, seven sick days without paperwork, and extra time to enjoy Christmas and New Year. Time to rest is part of the deal. Work that matters. Our mission is to build a digital world that is secure, accessible and inclusive for everyone. From fighting fraud to making online services easier and safer to use, your work will have a real impact on how people experience trust online. Compensation. We offer fair and transparent pay, benchmarked to the market. Truly global. We work across continents and time zones, with teammates and customers from all over the world. You will run campaigns that cross borders, cultures, and languages, and see your ideas land worldwide. Growth built in. Clear goals, open feedback and personal development plans. We support your progress with learning opportunities and by covering role-specific events, from design conferences to marketing forums. Celebrating your big moments. We are proud to support you through life’s biggest journeys with dedicated financial bonuses for both marriage and welcoming a new baby to your family. Team offsites. Sometimes just Slack is not enough. That is why we meet in person a few times a year. Trips are fully covered, so you can meet, collaborate, and recharge together. Getting you set up. We make sure you have access to the tools and hardware you need to do your work well. Friendly by design. Our logo is a dog for a reason. We keep things human, open and kind. We welcome individuality, quirks and different perspectives, because that is what makes our work smarter and more fun. The hiring stages: TA screening -> Hiring Manager Interview -> Technical Interview -> Final Interview. Sounds like a great opportunity for your career development? Then go ahead and apply! We are a global community of innovators, creators, and thinkers, and we believe that diversity fuels our innovation. Sumsub is proud to be an equal opportunity employer, committed to building a diverse and inclusive workforce. We welcome applications from people of all backgrounds, cultures, genders, experiences, abilities and perspectives. Join us in shaping the future inclusively. We believe in fair and transparent compensation. Salaries vary depending on work location and applicable local market conditions. This role is open to candidates in multiple countries, and compensation differs based on the candidate’s confirmed work location. For this reason, a single universal salary range is not included in this job posting. The applicable work location will be confirmed early in the recruitment process, and the relevant compensation information for that location will be shared in writing, in line with local requirements.

разрабатывать и дообучать модели TTS: проектирование, finetuning и адаптация генеративных подходов для высококачественного синтеза речи и zero shot клонирования голоса под задачи бизнеса; участвовать в дальнейшей разработке голосового пайплайна, объединяющего ASR, VAD, LLM, NLP и TTS для работы в режиме реального времени;