AI-инженер (LLM, агенты, AI product) вакансии

Задачи: — создать внутреннюю ИИ-систему для ювелирной компании — подключить базу 1С для возможности задавать вопросы в чате и получать информацию по товару: наличие, остатки, цена, местонахождение, количество — обеспечить поиск товара не только по артикулу, но и по фотографии, даже если фото отличается от изображения в 1С — развивать систему: подключить базу образов/луков, подбор похожих товаров и комплектов, рекомендации для клиентов, интеграцию с сайтом Требования: — специалист или команда с опытом в разработке ИИ-систем и интеграции с 1С Условия: — для отклика писать на @polcom07 с релевантными кейсами и примерной оценкой первого этапа

Задачи: — проектирование и реализация единой корпоративной агентной платформы, предоставляющей пользователю единую точку взаимодействия с информационными системами и бизнес-процессами; — разработка главного агента-оркестратора, специализированных агентов, реестра инструментов; — управление контекстом и состоянием, межагентное взаимодействие; — интеграция с корпоративными системами; — обеспечение безопасности, наблюдаемости, тестирование качества и отказоустойчивости; — hands-on участие в разработке. Требования: — от 3 лет коммерческой backend-разработки; — опыт технического проектирования распределённых систем; — практический опыт работы с LLM и агентными системами; — желательно наличие хотя бы одного агентного или LLM-продукта, доведённого до пилотной либо промышленной эксплуатации; — опыт технического лидерства или владения архитектурой крупного модуля; — умение работать в условиях неполных требований и формировать целевую архитектуру; — способность общаться с разработчиками и владельцами бизнес-процессов; — практический опыт разработки LLM-агентов и agentic workflows; — понимание различий между агентом, инструментом, workflow, multi-agent системой и чат-ботом; — опыт реализации supervisor/orchestrator-архитектуры; — планирование и декомпозиция сложных пользовательских задач; — выбор специализированного агента или инструмента; — реализация взаимодействия между агентами, параллельное выполнение независимых задач; — управление состоянием долгоживущих сценариев, работа с краткосрочной и долгосрочной памятью; — контроль циклов, timeout, retry и fallback; — разрешение конфликтующих результатов нескольких агентов; — реализация human-in-the-loop и запрос подтверждения перед критичными действиями; — защита от бесконтрольного поведения агентов; — межагентское взаимодействие: типизированные сообщения, контракты задач и результатов, делегирование задач, общий и изолированный контекст агентов, очереди и приоритеты задач, синхронное и асинхронное взаимодействие, передача управления, трассировка цепочки выполнения, предотвращение зацикливания, повторное выполнение и идемпотентность, обработка частичной недоступности агентов и инструментов; — сильное знание Python, async/await и конкурентное выполнение; — опыт работы с FastAPI, Pydantic, pytest; — проектирование API и типизированных контрактов; — модульная и гексагональная архитектура; — интеграция нового функционала в существующие системы; — опыт работы с REST, gRPC, WebSocket или Server-Sent Events; — работа с очередями сообщений и фоновыми задачами; — опыт работы с долгоживущими и асинхронными процессами; — понимание микросервисной архитектуры и модульных монолитов; — умение проектировать систему без жёсткой зависимости от конкретной LLM или фреймворка. Условия: — комфортный современный офис в Москве, ул. Вавилова 19; — офисный формат работы на период испытательного срока, после возможен гибрид; — ежегодный пересмотр зарплаты, структура дохода: оклад + 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; — 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.

Задачи: — Создание Telegram-бота "под ключ" — Интеграция с ИИ (ChatGPT или альтернатива) — Загрузка и адаптация текстового курса и базы знаний (объемная, структурированная информация) — Настройка корректной логики: пользователь → запрос → ИИ → ответ → бот отправляет пользователю — Обеспечение стабильной работы бота без VPN на российской или международной инфраструктуре Требования: — Опыт работы с ИИ-ботами (ChatGPT API, OpenAI, Mistral, Claude и другие) — Наличие примеров похожих реализованных проектов В отклике необходимо предоставить: — Примеры ИИ-ботов с доступом для просмотра (если есть) — Платформы и ИИ, с которыми работали (OpenAI, Hugging Face, Local LLM и пр.) — Сроки реализации — Стоимость проекта — Возможность сопровождения (если планируете)

Задачи: — Разбирать и обрабатывать документы в форматах Excel, PDF и сканы. — Извлекать табличные данные из документов. — Нормализовать даты, номера документов, суммы и другие поля. — Проектировать и реализовывать логику сопоставления данных. — Выявлять расхождения между двумя источниками. — Обрабатывать сложные кейсы: разные периоды, отличающиеся номера, частичные совпадения и другие неоднозначные ситуации. — Интегрировать OCR и Document AI-инструменты. — При необходимости использовать LLM для интерпретации и формирования человекочитаемых объяснений. — Разрабатывать backend/API для сервиса. — Участвовать в проектировании архитектуры MVP. — Работать с реальными обезличенными финансовыми документами и тестировать качество решения на исторических данных. Требования: — Коммерческий опыт разработки на Python от 3 лет. — Уверенное владение Python. — Опыт работы с pandas и табличными данными. — Опыт обработки Excel / CSV. — Опыт парсинга PDF. — Опыт разработки backend-сервисов и REST API. — Понимание алгоритмов сопоставления, нормализации и очистки данных. — Умение проектировать решение самостоятельно, а не только работать по готовому ТЗ. — Опыт работы с Git. Будет большим плюсом: — Опыт с OCR: Tesseract, PaddleOCR или аналогами. — Опыт с PyMuPDF, pdfplumber, Camelot, Tabula. — Опыт с fuzzy matching. — Опыт работы с Document AI / Intelligent Document Processing. — Интеграция LLM API: OpenAI, Azure OpenAI, YandexGPT или аналогов. — Опыт разработки систем reconciliation / data matching. — Опыт работы с финансовыми, бухгалтерскими или другими структурированными документами. — Опыт On-Premise решений. — Опыт построения решений, где нужно сочетать детерминированную бизнес-логику и AI. Что здесь не требуется: — Классический Data Scientist и обучение собственных ML-моделей с нуля. — В проекте важнее: сильная прикладная Python-разработка; работа с документами и данными; построение надежной логики сопоставления; понимание, где действительно нужен AI, а где лучше использовать обычные алгоритмы. О проекте: — На первом этапе планируется разработка MVP без глубокой интеграции с внутренними учетными системами. — Базовый сценарий: пользователь загружает акт сверки и внутреннюю выгрузку, система извлекает и нормализует данные, выполняет сопоставление, показывает совпадения и расхождения, формирует структурированный результат для пользователя. — Ожидаемый результат — рабочий прототип, который можно проверить на реальных исторических данных. Условия: — Удалённая работа. — Проектная занятость / аутстафф. — Возможна полная или частичная загрузка — обсуждается индивидуально. — Проект стартует с этапа MVP, с возможностью продолжения. — Работа в составе небольшой проектной команды. — Ожидаемый уровень Middle+ / Senior. — Специалист должен самостоятельно разобраться в предметной области, предложить техническое решение и довести его до рабочего результата. Будет преимуществом в отклике: — Опыт похожих задач: обработка документов; OCR; reconciliation / matching; financial data processing; Document AI; LLM-интеграции. — Указание текущей доступности, возможной загрузки в часах в месяц, часовой ставки, формата работы: ИП / самозанятый / другой.

Задачи: — Построить с нуля переиспользуемое ядро продукта: документы → граф знаний → гибридный поиск → ответ со ссылкой на пункт и статусом редакции. — Отвечать за архитектуру и разработку. — Первая цель — MVP. Обязанности: — Архитектура ядра. — Работа с данными. — Резолвинг перекрёстных ссылок, версионность и статусы актуальности документов. — Hybrid retrieval. — Graph RAG. — Evaluation. — Поставка. — Техлидерство. Требования: — 5+ лет backend-разработки на Python, из них 2+ года с LLM/RAG в продакшене (в идеале). — Графовые базы данных, моделирование онтологий. — Векторные базы данных, гибридный поиск, реранкинг. — LLM. — Опыт поставки в закрытые контуры. — Опыт руководства. Плюсом: — GraphRAG, entity extraction/resolution. — Внедрения в промышленности или проектных институтах. — React/TypeScript для интерфейса и админки. Условия: — Удалённая работа по России. — Гибкий график: 5/2 с 10:00 до 19:00 по Москве, с возможностью подстроить часы работы. — Клубы по интересам, летний выезд на природу AGIMA Camp, корпоративные вечеринки с возможностью приглашать семью. — Поддержка в создании личного бренда для профессионального роста. — Бесплатные билеты на забег для IT-специалистов RUNIT. — ДМС со стоматологией. — Оформление по ГПХ с физлицом.

Задачи: — построить AI-стек для инженерии; — внедрить AI-агентов, evals, AI code review и автоматизацию тестов/деплоя; — ускорить разработку в 5 раз; — выбирать AI-модели и инструменты; — запускать MVP и масштабировать успешные решения на всю компанию. Требования: — 6–7+ лет коммерческой разработки; — опыт масштабного внедрения AI в инженерные процессы; — Claude Code, Cursor и аналогичные инструменты; — опыт с AI-агентами; — data-driven подход; — английский B2+. Условия: — работа в международной B2C-компании в сфере развлечений; — локация Кипр / UK, remote из Европы; — full-time / contract; — релокация и visa support.

Задачи: — Проектировать, создавать и поддерживать сложные рабочие процессы в n8n для автоматизации маркетинговых процессов. — Заменять ручные и повторяющиеся маркетинговые задачи надежной и масштабируемой автоматизацией. — Обеспечивать надежность, документацию и удобство поддержки рабочих процессов. — Создавать AI-агентов для поддержки маркетинговой отчетности, генерации инсайтов, создания контента и экспериментов. — Проектировать логику запросов, цепочки и потоки решений для получения последовательных и полезных результатов. — Постоянно улучшать AI-агентов на основе обратной связи и производительности. — Автоматизировать сбор, анализ и суммирование маркетинговых данных по каналам. — Создавать автоматизированные системы отчетности по пользователям, пробным периодам, подпискам, доходам и эффективности воронки. — Повышать скорость получения инсайтов с помощью AI-генерируемых сводок и анализа. — Выявлять узкие места и неэффективности в маркетинговой команде и предлагать решения по автоматизации. — Обеспечивать более быстрое выполнение задач и принятие решений через хорошо спроектированные рабочие процессы и инструменты. — Поддерживать команды по росту, производительности, SEO и контенту с помощью индивидуальных решений по автоматизации. — Тесно сотрудничать с заинтересованными сторонами маркетинга для понимания их потребностей и приоритизации возможностей автоматизации. — Взаимодействовать с командами продукта, данных и инженерии при необходимости. — Быстро итеративно улучшать рабочие процессы и AI-агентов по мере изменения требований. Требования: — Подтвержденный практический опыт создания сложных рабочих процессов в n8n. — Глубокое понимание логики рабочих процессов, триггеров, вебхуков, API и трансформации данных. — Опыт интеграции AI или больших языковых моделей в рабочие процессы для анализа, суммирования или генерации контента. — Сильные навыки решения проблем и способность переводить бизнес-требования в технические решения. — Знакомство с маркетинговыми данными, воронками и отчетностью в SaaS-среде. — Четкие письменные и устные коммуникативные навыки на английском языке. — Внимание к деталям и фокус на надежности и поддерживаемости. Желательно: — Опыт работы в командах маркетинга, роста или аналитики. — Опыт работы с инструментами GA4, рекламными платформами, CRM или хранилищами данных. — Опыт создания внутренних инструментов или фреймворков автоматизации. — Опыт работы в B2B SaaS-среде. Условия: — Удаленная работа с полной занятостью. — Возможность влиять на работу маркетинга в масштабе, улучшая производительность, принятие решений и качество результатов. — Работа на пересечении AI, автоматизации и роста с возможностью экспериментировать, итеративно улучшать и внедрять значимые решения.

Location: EMEA; United States; Ireland; United Kingdom; Brazil; Canada; Mexico ✨ ONE PLATFORM, A WHOLE WORLD OF OPPORTUNITY The best jobs have always clustered in a handful of the world's wealthiest cities. But talent is everywhere. Oyster set out to close that gap - building a global employment platform that lets companies hire, pay, and care for brilliant people anywhere. We're proof that a high-performing culture doesn't need an office. Distributed across 60+ countries since 2020, we've built something the industry keeps noticing: - Ranked #10 of 250 on TIME and Statista's 2026 list of America's Top WorkTech Companies https://www.oysterhr.com/library/oyster-hr-named-10-of-americas-top-worktech-companies-of-2026-by-time - Named one of America's Greatest Startup Workplaces 2026 https://www.oysterhr.com/library/oyster-2026-newsweek-award by Newsweek - A G2 Spring 2026 Leader https://www.oysterhr.com/library/oyster-named-best-roi-for-any-global-employment-platform-for-g2-spring-2026 across Employer of Record, Global Employment, Multi-country Payroll, and HR Compliance - The only B Corp-certified global employment platform ~ independently verified since 2023 Two of those rankings measured our business impact from the outside. One measured how our own people feel about working here. They landed in the same place ~ because at Oyster, culture and performance aren't separate conversations. They're the same one. And we're just getting started. If you want to do the best work of your career alongside people who care as much as you do, we'd love for you to apply. Location: Fully remote. To create the best experience for you and to meet our business needs, this role requires you to be based within the EMEA or AMER regions. You’ll join Oyster as a Senior AI Solutions Engineer: Go-to-Market, reporting to the AI Director and partnering closely with GTM leadership. Your mandate goes beyond automating individual tasks. You’ll help make Sales, Marketing, Customer Success, Partnerships, and Revenue Operations AI native. This means rethinking how people work with AI, where agents can take on operational work, and where human judgement creates the most value to build a more effective AI enabled GTM organization. You’ll be part of a connected group of Senior AI Solutions Engineers transforming different departments across Oyster. Together, you’ll establish shared standards, reuse successful approaches, and ensure Oyster’s AI transformation is coordinated rather than siloed. This is a senior, hands on role. You’ll independently identify high value opportunities, decide where AI should and should not be used, and lead complex solutions from discovery through adoption and measurable business impact.   CORE RESPONSIBILITIES Make GTM AI native Reimagine how GTM teams work with AI. Move teams from manually operating tools and workflows toward humans orchestrating AI powered ways of working, with human judgement where it matters. Shape the AI roadmap Partner with GTM leaders to identify, prioritize, and frame high impact transformation opportunities. Lead solutions from start to finish Own complex initiatives from an ambiguous problem through design, build, deployment, adoption, and improvement. Build best in class AI agents and solutions Design, build, test, and continuously improve reliable AI agents, agentic workflows, automations, and integrations using LLMs, orchestration platforms, APIs, tools, and enterprise data sources. Apply sound judgement Balance value, feasibility, risk, data quality, maintainability, and human oversight. Recognize when AI is not the right solution. Drive AI adoption and self sufficiency Partner with GTM leaders, managers, and champions to embed AI into everyday work. Help teams move beyond consuming AI outputs toward confidently creating reusable prompts, workflows, and AI powered ways of working. Own solutions beyond launch Measure adoption, quality, reliability, and business impact after release. Continuously improve solutions based on real usage and outcomes rather than treating launch as completion. Connect the organization and raise capability Build trust with GTM leadership, share reusable patterns with AI Solutions Engineers supporting other departments, mentor colleagues, and help teams develop the skills and confidence to operate effectively in an AI enabled environment.   MINIMUM REQUIREMENTS - 3+ years of hands on AI and automation experience, supported by broader experience delivering complex solutions in GTM Engineering, Solutions Engineering, Revenue Operations, systems integration, or a related field. - A track record of redesigning workflows or operating models, rather than only automating isolated tasks. - Experience independently leading complex, cross functional initiatives from an ambiguous problem to measurable business impact. - Strong automation capability using tools such as n8n, Zapier, Make, or comparable orchestration technologies. - Strong hands on experience designing, building, deploying, and improving production AI agents and agentic workflows using LLMs, APIs, tools, and enterprise data. - Strong understanding of B2B GTM operations across Sales, Marketing, Customer Success, Partnerships, and Revenue Operations. - Experience integrating business systems through APIs, webhooks, and authentication mechanisms. - Strong judgement about where AI can create value, where processes need deeper redesign, and where automation should not be used. - Ability to influence functional leaders, challenge assumptions constructively, and align technical and nontechnical stakeholders. - A track record of mentoring others, sharing expertise, and raising standards across a team.   BONUS EXPERIENCE - Experience with Salesforce, HubSpot, Gong, Clay, or similar GTM platforms. - JavaScript, Python, or SQL experience. - Familiarity with RAG, MCP, AI evaluation, observability, governance, or security controls. HOW WE WORK TOGETHER AT OYSTER Our values guide the work we do, the decisions we make, and the culture that makes Oyster what it is. We make it happen—progress is our default setting. We drive change, we build and give trust, and we move as one united team. Our mission is to create a more equal world of work: helping companies everywhere hire, pay, and care for talent anywhere. Everything we do ladders up to our mission. We embrace asynchronous, collaborative work and share how we operate in the Oyster Public HQ https://oysterhr.notion.site/Welcome-to-The-Reef-Oyster-s-Open-Source-Employee-Guide-fc482e3f107c4afd8c85ed217f1d0c1d so other global teams can learn from us. HOW YOU WORK With diverse locations, cultures, and needs, we created How YOU Work; https://oysterhr.notion.site/How-YOU-work-ac16c1932e314ffd8cb693e332458395 a program supporting your whole human experience at Oyster: - Work from anywhere: Oyster has no borders or HQ. As long as work is timely, your team is supported, and you're authorized to work where you live, you can work from anywhere. - Paid time off: Enjoy 40 days off per year (including holidays and vacation), or more if required by your country. - Mental health support: Access Plumm https://oysterhr.notion.site/Plumm-3b5e7e70dec6428e975f68f25972cfb7, our mental well-being service. - Wellbeing allowance: Each month, receive a wellbeing allowance in your ThanksBen wallet to spend on a wide range of options. - Flexible parental leave: All new parents are eligible for at least three months’ paid leave, with job protection for up to 12 months or as required locally. - WFH stipend: Receive a stipend for your laptop and home office equipment to get you set up quickly. ✨ THE BEST JOBS SHOULD BE AVAILABLE TO EVERYONE  We embrace diverse perspectives and intentionally build our product and company with an inclusive, global mindset. We encourage you to apply, even if you don’t meet every criterion, and are committed to a trust-based, thriving environment. Oyster is committed to ensuring equal opportunity of employment for qualified persons with disabilities and actively fosters an inclusive work environment. If you require reasonable accommodations throughout the recruiting process, please email jobs@oysterhr.com with the subject line: Interview Accommodations. Our available positions are on our careers page. Our team will only contact you from an @oysterhr.com email, and we will never ask for money as part of an interview process or job offer. If you receive a suspicious email about Oyster jobs or are directed to a site other than www.oysterhr.com/careers http://www.oysterhr.com/careers, please report it via our Compliance and Ethics Helpline https://faceup.com/c/8tpzyyye.

Location: Lithuania OUR HIRING PROCESS: We will review your application against our job requirements. We do not employ machine learning technologies during this phase as we believe every human deserves attention from another human. We do not think machines can evaluate your application quite like our seasoned recruiting professionals—every person is unique. We promise to give your candidacy a fair and detailed assessment. We may then invite you to submit a video interview for the review of the hiring manager. This video interview is often followed by a test or short project that allows us to determine whether you will be a good fit for the team. At this point, we will invite you to interview with our hiring manager and/or the interview team. Please note: We do not conduct interviews via text message, Telegram, etc. and we never hire anyone into our organization without having met you face-to-face (or via Zoom). You will be invited to come to a live meeting or Zoom, where you will meet our INFUSE team. From there on, it’s decision time! If you are still excited to join INFUSE and we like you as much, we will have a conversation about your offer. We do not make offers without giving you the opportunity to speak with us live. INFUSE is committed to complying with applicable data privacy and security laws and regulations. For more information, please see our Privacy Policy We're looking for Senior AI Engineers to design, build and run LLM-powered agents and AI features in production. You'll work across the full lifecycle: from scoping a use case with stakeholders, through architecture and development, to deployment, evaluation and ongoing improvement. This is a hands-on role for engineers who think like product owners: you understand the business problem, define what success looks like in numbers, ship to real users, and own the result after launch. Our stack: Python and TypeScript/Node.js, PostgreSQL, Kafka, Docker, GitHub Actions, AWS. We run multiple LLM providers (Anthropic, OpenAI, Google, AWS Bedrock and others) behind a shared LLM gateway, with MCP for tool access. We work AI-native: agentic coding tools are part of how engineering gets done here, and we provide official Anthropic Claude certification for our engineers. As part of our team, your responsibilities in this role will include: Design and build LLM agents and agentic workflows: tool use via MCP, structured outputs, retrieval (RAG), multi-step orchestration, human review steps where needed. Build and maintain MCP servers and tool layers that give agents governed access to company data and systems. Own quality and cost: evals and golden sets, per-agent telemetry (tokens, cost, latency, accuracy), prompt and model versioning, provider failover and model routing. Ship through CI/CD like any other production service: tests and evals in the pipeline, containerized deploys, monitoring, rollbacks. Contribute to shared platform pieces: the agent template, the LLM gateway, observability. Work directly with business stakeholders to scope, prioritize and validate use cases. Push back when something doesn't need an agent. Use AI coding tools daily. Claude Code is our default; deep experience with a comparable agentic tool also works. You drive the tool, verify its output and catch its mistakes. Our ideal candidate possesses the following skills: 5+ years of software engineering experience, with production systems you've designed, shipped and supported. Proven experience building LLM-based agents or agentic workflows that reached production with real users. Hands-on experience with agent frameworks and SDKs: Claude Agent SDK, Claude Managed Agents, Pydantic AI, LangGraph or similar. You can also explain what you'd build without one. Hands-on experience with MCP and RAG: you've built MCP servers or equivalent tool layers, and retrieval pipelines over real, messy data. Experience with multiple LLM providers and with AWS Bedrock: model selection, routing, fallbacks, cost and rate-limit management. Strong Python and SQL. Comfortable in TypeScript/Node.js. Solid production engineering on AWS: Docker, CI/CD, testing, observability. Product thinking and business focus: you start from the problem and its cost, define success in measurable terms, and own outcomes. Daily use of agentic coding tools (Claude Code, Codex, Cursor or similar), with the judgment to verify what they produce. Clear communication with technical and non-technical stakeholders, in English. Nice to have: LLM observability or gateway tooling (Langfuse, LiteLLM, OpenTelemetry for LLM traces or similar). B2B contact and company data: enrichment, deduplication, matching. Cloud operations or FinOps: monitoring, incident handling, cost management. Event-driven or workflow-engine experience (Kafka, Temporal or similar). Experience replacing no-code or workflow-tool automations with production code. What we offer: Fully remote work format; A collaborative and creative environment that values initiative and experimentation; Paid time off and sick leave; Reduced working hours on Fridays during the summer; Days-off on US national holidays + your birthday; Schedule: 14 PM - 23 PM EEST (Kyiv time).

Location: Romania OUR HIRING PROCESS: We will review your application against our job requirements. We do not employ machine learning technologies during this phase as we believe every human deserves attention from another human. We do not think machines can evaluate your application quite like our seasoned recruiting professionals—every person is unique. We promise to give your candidacy a fair and detailed assessment. We may then invite you to submit a video interview for the review of the hiring manager. This video interview is often followed by a test or short project that allows us to determine whether you will be a good fit for the team. At this point, we will invite you to interview with our hiring manager and/or the interview team. Please note: We do not conduct interviews via text message, Telegram, etc. and we never hire anyone into our organization without having met you face-to-face (or via Zoom). You will be invited to come to a live meeting or Zoom, where you will meet our INFUSE team. From there on, it’s decision time! If you are still excited to join INFUSE and we like you as much, we will have a conversation about your offer. We do not make offers without giving you the opportunity to speak with us live. INFUSE is committed to complying with applicable data privacy and security laws and regulations. For more information, please see our Privacy Policy We're looking for Senior AI Engineers to design, build and run LLM-powered agents and AI features in production. You'll work across the full lifecycle: from scoping a use case with stakeholders, through architecture and development, to deployment, evaluation and ongoing improvement. This is a hands-on role for engineers who think like product owners: you understand the business problem, define what success looks like in numbers, ship to real users, and own the result after launch. Our stack: Python and TypeScript/Node.js, PostgreSQL, Kafka, Docker, GitHub Actions, AWS. We run multiple LLM providers (Anthropic, OpenAI, Google, AWS Bedrock and others) behind a shared LLM gateway, with MCP for tool access. We work AI-native: agentic coding tools are part of how engineering gets done here, and we provide official Anthropic Claude certification for our engineers. As part of our team, your responsibilities in this role will include: Design and build LLM agents and agentic workflows: tool use via MCP, structured outputs, retrieval (RAG), multi-step orchestration, human review steps where needed. Build and maintain MCP servers and tool layers that give agents governed access to company data and systems. Own quality and cost: evals and golden sets, per-agent telemetry (tokens, cost, latency, accuracy), prompt and model versioning, provider failover and model routing. Ship through CI/CD like any other production service: tests and evals in the pipeline, containerized deploys, monitoring, rollbacks. Contribute to shared platform pieces: the agent template, the LLM gateway, observability. Work directly with business stakeholders to scope, prioritize and validate use cases. Push back when something doesn't need an agent. Use AI coding tools daily. Claude Code is our default; deep experience with a comparable agentic tool also works. You drive the tool, verify its output and catch its mistakes. Our ideal candidate possesses the following skills: 5+ years of software engineering experience, with production systems you've designed, shipped and supported. Proven experience building LLM-based agents or agentic workflows that reached production with real users. Hands-on experience with agent frameworks and SDKs: Claude Agent SDK, Claude Managed Agents, Pydantic AI, LangGraph or similar. You can also explain what you'd build without one. Hands-on experience with MCP and RAG: you've built MCP servers or equivalent tool layers, and retrieval pipelines over real, messy data. Experience with multiple LLM providers and with AWS Bedrock: model selection, routing, fallbacks, cost and rate-limit management. Strong Python and SQL. Comfortable in TypeScript/Node.js. Solid production engineering on AWS: Docker, CI/CD, testing, observability. Product thinking and business focus: you start from the problem and its cost, define success in measurable terms, and own outcomes. Daily use of agentic coding tools (Claude Code, Codex, Cursor or similar), with the judgment to verify what they produce. Clear communication with technical and non-technical stakeholders, in English. Nice to have: LLM observability or gateway tooling (Langfuse, LiteLLM, OpenTelemetry for LLM traces or similar). B2B contact and company data: enrichment, deduplication, matching. Cloud operations or FinOps: monitoring, incident handling, cost management. Event-driven or workflow-engine experience (Kafka, Temporal or similar). Experience replacing no-code or workflow-tool automations with production code. What we offer: Fully remote work format; A collaborative and creative environment that values initiative and experimentation; Paid time off and sick leave; Reduced working hours on Fridays during the summer; Days-off on US national holidays + your birthday; Schedule: 14 PM - 23 PM EEST (Kyiv time).

Компания: международный B2C бизнес в индустрии развлечений, работающий с несколькими продуктами для музыкантов и авторов по всему миру. Задачи: — Построить и возглавить AI-стек для инженерии: агентные воркфлоу, evals, правила доступа, AI-ревью кода, автоматизацию тестирования и деплоя — Находить и убирать главные узкие места в рабочих процессах с помощью AI — Отвечать за 5-кратное ускорение разработки — Выбирать модели и провайдеров для инструментов — Запускать MVP на одной команде и масштабировать на всю организацию — Настраивать evals, которые показывают реальный эффект на скорость Требования: — 6-7+ лет коммерческой разработки, любой стек, последние несколько лет фокус на внедрении и автоматизации процессов через AI — Опыт внедрения AI-инструментов в масштабе всей компании с измеримым эффектом на скорость работы — Экспериментальный подход, основанный на данных: предложить идею, запустить MVP на одной команде, затем масштабировать то, что сработало — Опыт работы с агентными инструментами (Claude Code, Cursor и подобные) в масштабе организации, а не только личное использование — Английский от B2 Условия: — Официальное оформление на Кипре или в UK — Для remote - b2b контракт — Для кандидатов из РФ/РБ есть поддержка релокации на Кипр — Remote возможен, чем ближе к часовому поясу CET, тем лучше — Релокация + visa support — Обучение, языковые занятия, конференции — Поддержка ментального здоровья

Location: Lithuania; Bosnia & Herzogovina; North Macedonia; Romania; Slovakia; Slovenia; Serbia; Malta; Moldova; Montenegro; Portugal; Estonia; Latvia; Albania; United Kingdom; Czech Republic; Greece; Cyprus; Kosovo; Bulgaria; Poland; Georgia; Bosnia and Herzegovina; Turkey; Croatia; South Africa Workplace: Remote [https://app.ashbyhq.com/api/images/user-content/b540b9e2-0673-4d2d-91fd-e293658f4f6a/b7561393-0797-4849-9149-363b09bab85c/job%20ad%20visual.png] Please be informed that Foxelli Group hires exclusively in-house and doesn't work with any external recruitment agencies. Any company or individual claiming to recruit on behalf of Foxelli Group is not affiliated with us. At Foxelli, we’re building more than e-commerce brands - we’re building a tribe of thoughtful humans: AI enthusiasts, curious builders, and ambitious minds who care about doing meaningful work together. Ten years in, our D2C brands reach millions and generate over $20M annually. Our focus stays simple: create products customers love and a team that actually enjoys building them. If you love to learn, experiment, think boldly, make decisions, and build things that matter - you’ll feel at home with us. WHY JOIN FOXELLI? Foxelli is a place for people who think deeply, learn quickly, automate repetitive work, enjoy spending time together, and build things that matter. We work hard, joke often, and don’t take ourselves too seriously. If you’re excited about AI, eager to learn, and want to create work that genuinely means something with people who care, you’ll feel right at home with us - no matter where you are in your career. What You'll Do We create premium AI short films and series that make people feel something. You'll be a key part of building this, producing daily episodes of dark romance, horror, thriller, and fantasy micro-dramas for TikTok, Instagram Reels, and YouTube Shorts. Each episode is 60-90 seconds of polished, character-driven storytelling that lives on someone's phone, often watched without sound. You'll work at the intersection of cutting-edge AI tools and deliberate creative craft. This means: - Striking visuals that stop the scroll and draw the eye - Believable characters who feel real, episode to episode - Natural-sounding voices and carefully placed audio that works whether the sound is on or off - Purposeful editing where every cut serves the story and the viewer's emotional journey - Style in service of story—the look and feel should match the world and tone of the series You'll move fast, but never at the expense of craft. Every frame should feel intentional What We're Looking For - 2-5 years of professional experience working with AI tools for video, image, or audio generation - Solid understanding of editing fundamentals: frame composition, pacing, visual flow, and how cuts guide audience attention - Prompt engineering skills -ability to write precise, iterative prompts that yield the exact creative output you need - Portfolio demonstrating completed AI-generated video work or traditional video edited to a high standard - Ability to work independently and ship fast without constant creative direction - Familiarity with vertical video format requirements and platform-specific best practices It will be a plus: - Experience in classic video production (films, TV series, shows) - Understanding of dramaturgy and story construction - The full cycle of AI video creation - from idea to final - Experience working with short drama or creating AI content for TikTok / YouTube THE RECRUITMENT PROCESS 1. Apply - A simple form that helps us know who’s stepping into our world. 2. Text Task - A small real-world challenge so we can see how you approach problems in action. 3. Hire - if we align, we'll invite you to join our team for a test trial period. FIXED SERVICE FEE Up to 2500 euros per month after taxes (depending on your skills, competencies, and experience, we'd be happy to negotiate your salary). PERKS & BENEFITS - Full flexibility with your schedule in a fully remote setting, while Hubstaff ensures transparency and fairness. - Complimentary gym memberships and company-sponsored outdoor adventures for your strength and to keep your physical well-being in check. - A personal development budget is at your disposal for your personal and professional growth to ensure you keep growing continuously. - Generous paid time off, including parental leave and vacation days, to recharge and rejuvenate and ensure you’re well-rested. - A budget for your home office setup so you’re comfortable working how you like and where you like. - Strong company culture and a network of unique individuals that celebrate you. ELIGIBILITY & LOCATION REQUIREMENTS ❕This role is 100% remote. - Our tribe works primarily in the Eastern European Time Zone (EET — UTC+2/UTC+3). We also use Revolut for payments, so we are open to hiring candidates in countries where Revolut payments are supported. - For full-time roles, we ask that you’re available for at least 5 hours during typical EET working hours — it keeps collaboration smooth and communication easy. (This doesn’t apply to part-time roles.) Currently, we are hiring in Albania, Bosnia and Herzegovina, Croatia, Czech Republic, Estonia, Greece, Georgia, Kosovo, Latvia, Lithuania, Moldova, Montenegro, North Macedonia, Poland, Portugal, Romania, Slovakia, Malta, Slovenia, Serbia, Cyprus, Bulgaria, Hungary, Netherlands, the United Kingdom, and South Africa.

Лаборатория «Космос» — российский стартап, создающий собственные программно-аппаратные продукты: от первых прототипов до готовых устройств. В портфеле два продукта: «Каспер» — робо-компьютер для обучения детей; «Света» — интерактивная светозвуковая панель для дома и различных пространств. Задачи: — исследовать новые AI-модели, сервисы и способы их применения; — решать отдельные продуктовые задачи с помощью существующих AI-технологий: исследовать рынок, выбирать решение, тестировать и собирать прототип, например, найти и интегрировать технологию для работы Каспера с произношением ребёнка; — собирать внутренние AI-инструменты для команды из готовых моделей, API и автоматизаций; — автоматизировать работу с документами, голосовыми сообщениями, ссылками, PDF и базой знаний; — создавать инструменты для работы команды с видео, изображением, музыкой, звуком и другим контентом; — проводить исследования по продуктам, технологиям, конкурентам и патентам. Требования: — очень хорошее знание современных нейросетей, основных моделей и сервисов; — постоянное отслеживание новинок и тестирование; — умение применять AI-технологии для решения задач; — API, автоматизации и базовый код важны; — опыт классической разработки будет плюсом, но не обязателен; — самостоятельность, любопытство, способность быстро проходить путь от задачи и исследования до работающего решения. Условия: — работа в офисе в Санкт-Петербурге, центр города, м. Горьковская; — оплачиваемая работа; — занятость и график обсуждаются индивидуально; — уровень вознаграждения зависит от навыков, опыта и объёма занятости.

Location: Ukraine OUR HIRING PROCESS: We will review your application against our job requirements. We do not employ machine learning technologies during this phase as we believe every human deserves attention from another human. We do not think machines can evaluate your application quite like our seasoned recruiting professionals—every person is unique. We promise to give your candidacy a fair and detailed assessment. We may then invite you to submit a video interview for the review of the hiring manager. This video interview is often followed by a test or short project that allows us to determine whether you will be a good fit for the team. At this point, we will invite you to interview with our hiring manager and/or the interview team. Please note: We do not conduct interviews via text message, Telegram, etc. and we never hire anyone into our organization without having met you face-to-face (or via Zoom). You will be invited to come to a live meeting or Zoom, where you will meet our INFUSE team. From there on, it’s decision time! If you are still excited to join INFUSE and we like you as much, we will have a conversation about your offer. We do not make offers without giving you the opportunity to speak with us live. INFUSE is committed to complying with applicable data privacy and security laws and regulations. For more information, please see our Privacy Policy We're looking for Senior AI Engineers to design, build and run LLM-powered agents and AI features in production. You'll work across the full lifecycle: from scoping a use case with stakeholders, through architecture and development, to deployment, evaluation and ongoing improvement. This is a hands-on role for engineers who think like product owners: you understand the business problem, define what success looks like in numbers, ship to real users, and own the result after launch. Our stack: Python and TypeScript/Node.js, PostgreSQL, Kafka, Docker, GitHub Actions, AWS. We run multiple LLM providers (Anthropic, OpenAI, Google, AWS Bedrock and others) behind a shared LLM gateway, with MCP for tool access. We work AI-native: agentic coding tools are part of how engineering gets done here, and we provide official Anthropic Claude certification for our engineers. As part of our team, your responsibilities in this role will include: Design and build LLM agents and agentic workflows: tool use via MCP, structured outputs, retrieval (RAG), multi-step orchestration, human review steps where needed. Build and maintain MCP servers and tool layers that give agents governed access to company data and systems. Own quality and cost: evals and golden sets, per-agent telemetry (tokens, cost, latency, accuracy), prompt and model versioning, provider failover and model routing. Ship through CI/CD like any other production service: tests and evals in the pipeline, containerized deploys, monitoring, rollbacks. Contribute to shared platform pieces: the agent template, the LLM gateway, observability. Work directly with business stakeholders to scope, prioritize and validate use cases. Push back when something doesn't need an agent. Use AI coding tools daily. Claude Code is our default; deep experience with a comparable agentic tool also works. You drive the tool, verify its output and catch its mistakes. Our ideal candidate possesses the following skills: 5+ years of software engineering experience, with production systems you've designed, shipped and supported. Proven experience building LLM-based agents or agentic workflows that reached production with real users. Hands-on experience with agent frameworks and SDKs: Claude Agent SDK, Claude Managed Agents, Pydantic AI, LangGraph or similar. You can also explain what you'd build without one. Hands-on experience with MCP and RAG: you've built MCP servers or equivalent tool layers, and retrieval pipelines over real, messy data. Experience with multiple LLM providers and with AWS Bedrock: model selection, routing, fallbacks, cost and rate-limit management. Strong Python and SQL. Comfortable in TypeScript/Node.js. Solid production engineering on AWS: Docker, CI/CD, testing, observability. Product thinking and business focus: you start from the problem and its cost, define success in measurable terms, and own outcomes. Daily use of agentic coding tools (Claude Code, Codex, Cursor or similar), with the judgment to verify what they produce. Clear communication with technical and non-technical stakeholders, in English. Nice to have: LLM observability or gateway tooling (Langfuse, LiteLLM, OpenTelemetry for LLM traces or similar). B2B contact and company data: enrichment, deduplication, matching. Cloud operations or FinOps: monitoring, incident handling, cost management. Event-driven or workflow-engine experience (Kafka, Temporal or similar). Experience replacing no-code or workflow-tool automations with production code. What we offer: Fully remote work format; A collaborative and creative environment that values initiative and experimentation; Paid time off and sick leave; Reduced working hours on Fridays during the summer; Days-off on US national holidays + your birthday; Schedule: 14 PM - 23 PM EEST (Kyiv time).

Location: United States Workplace: Remote BUILDING THE FUTURE OF OPEN FINANCE Payward - the parent company behind Kraken, NinjaTrader, Breakout, xStocks, Payward Services and CF Benchmarks - has spent the last 15 years building one of the most modern and globally accessible financial infrastructure platforms in the industry, built to advance an open, global financial system. THE TEAM Founded in 2011, Kraken is one of the world's longest-standing crypto platforms, trusted by over 10 million individuals and institutions across the globe. It offers spot trading, margin, futures, staking, and OTC services, with products built for both individual investors and institutional clients. Enterprise Transformation owns Kraken's internal AI platform. AI tooling is deployed at scale across the org. Tiered spend controls are live, a governed integration gateway is in production, and a system inventory is being built against real regulatory obligations. The next constraint is not access. It's capability. The frontier moves every few weeks, and most of what arrives is a protocol, API, or primitive that someone has to turn into something the rest of the company can safely use.   THE OPPORTUNITY - Design and build the technical layer of the AI platform, from integration surface to production runbook. Identify capability gaps between what's possible and what the organization can safely use, then close them by evaluating vendors or building internally. - Own the governed integration gateway as an engineering surface. You'll handle connector onboarding, authentication flows including OAuth and federated identity, reliability, and vendor escalation when things break. - Build internal tooling that extends the platform. Think agent scaffolding, reusable skills and prompt assets, evaluation harnesses for model changes, provisioning automation, and integrations between AI tooling and your systems of record. - Configure and operate tiered spend controls across the AI estate. Set caps by role and tier, build alerting rules, manage exception queues, run monthly reconciliation against vendor commitments, and eliminate the current single author risk in cap automation. - Implement the technical controls behind policy on data handling, retention, and model eligibility. This means identity integration, SSO and SCIM, entitlement by role and tier, and workspace configuration where retention terms differ by model class. - Work directly with teams to unblock them. Sit with them, understand the workflow, build or configure the thing that enables them to get real leverage from the tools. - Run technical enablement. Deep dives for tool rollouts, office hours for model releases, and authority over the engineering detail in knowledge base material. - Administer the enterprise AI estate end to end. Model turn ups, feature enablement, deprecation, provisioning queue, vendor configurations, and seat true ups.   WHAT YOU BRING - Enterprise SaaS administration at 1,000+ seats, ideally including an AI or LLM platform. - Demonstrated ability to build and ship working software, not just configure vendor products. Python or a comparable language should be natural to you. Version control, CI, containers, and secrets management are everyday tools. - Experience integrating against LLM and SaaS APIs with working familiarity of agent and tool use patterns, including MCP or equivalent tool calling architectures. - Working knowledge of identity and access. Okta or an equivalent IdP, OAuth, SCIM, SSO, and federated machine identity are things you've worked with. - Operational discipline. You understand intake queues, SLAs, runbooks, escalation hygiene, and the instinct to document a fix the first time you make it. - Genuine teaching ability. You can explain a technical capability three different ways on the same day to compliance, traders, and backend engineers. - Clear written communication for technical documentation, runbooks, and policy pages. - Judgment about what to build. You can tell the difference between a capability the organization actually needs and a demo that will be abandoned in a month.   NICE TO HAVES - Experience standing up an internal developer or AI platform, including the self service and guardrail layers. - Exposure to FinOps or software asset management practice. - Experience with low code or workflow automation platforms. - Atlassian administration depth in Jira, Confluence, or JSM. - Familiarity with regulated industry audit and evidence work in financial services. Unless a specific application deadline is stated in the job posting, applications are accepted on an ongoing basis. Please note, applicants are permitted to redact or remove information on their resume that identifies age, date of birth, or dates of attendance at or graduation from an educational institution. We consider qualified applicants with criminal histories for employment on our team, assessing candidates in a manner consistent with the requirements of the San Francisco Fair Chance Ordinance. OUR COMMITMENT Payward is powered by people from around the world and we celebrate the diverse talents, backgrounds, contributions, and unique perspectives that everyone brings to the table. We hire based on merit, seeking out people with the right abilities, knowledge, and skills for the job. We encourage you to apply for roles where you don't fully meet the listed requirements, especially if you're passionate or knowledgeable about crypto. We may ask candidates to complete job-related skills or work-style assessments as part of our hiring process. These assessments evaluate competencies relevant to the role and are applied consistently across candidates for similar positions. Results are considered alongside experience and interviews, and are not the sole basis for any employment decision. As an equal opportunity employer, we don't tolerate discrimination or harassment of any kind, whether based on race, ethnicity, age, gender identity, citizenship, religion, sexual orientation, disability, pregnancy, veteran status, or any other protected characteristic as outlined by federal, state, or local laws. Stay connected Follow us on Twitter https://twitter.com/krakenfx Learn on the Kraken Blog https://blog.kraken.com/#:~:text=Enter%20your%20email%20address Connect on LinkedIn https://www.linkedin.com/company/kraken-exchange/ Candidate Privacy Notice https://www.kraken.com/legal/candidate-privacy-notice