Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$94к$144кмедиана$222к
10 в выборке
На СНГ рынке: $6к/мес ($3.9к - $7.5к)
92
Отличная вакансия
развернуть
Это чётко определённая техническая лидерская роль на передовых системах agentic RL с современным стеком и явными сигналами хорошего баланса работы и жизни; уровень оплаты при этом не раскрыт.
Передовые технологии
Понятная зона ответственности
Сильный фокус на ИИ
Гибкие условия
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Lead Research Engineer (Agentic RL): Architecting and optimizing infrastructure for frontier-scale agentic LLM reinforcement learning, including asynchronous training, MoE models above 300 billion parameters, long-context interactions, and tool-using agent harnesses with an accent on distributed systems, numerical consistency, and fault tolerance. Focus on profiling the full RL loop, building reproducible training pipelines, solving distributed and performance issues, and leading engineering standards for research at scale.
Location: Hybrid in Zug, Switzerland, or London, United Kingdom
Company
hirify.global provides trusted content, software, and insights for professionals and institutions across legal, tax, accounting, compliance, government, and media.
What you will do
Own the architecture and technical direction of large-scale reinforcement learning training infrastructure for agentic LLMs and MoE models.
Optimize the full RL loop, including generation, tool calls, reward and judge computation, training, and weight synchronization.
Build asynchronous, fault-tolerant pipelines for models interacting with tools and agent harnesses.
Diagnose numerical, distributed, and performance issues across training frameworks, inference engines, and model code.
Partner with Research Scientists to test algorithms, rewards, and environments quickly and at scale.
Set standards for code quality, testing, reproducibility, and mentorship.
Requirements
MSc or PhD in computer science, machine learning, or a related field, or equivalent industry experience.
3+ years of professional ML engineering or research engineering experience in large-scale model training.
Extensive experience training large models on multi-node GPU clusters with advanced PyTorch and Megatron-LM/Megatron-Core, FSDP/DTensor, or DeepSpeed.
Deep understanding of tensor, pipeline, context, and expert parallelism, including memory and communication trade-offs for MoE and long-context models.
Hands-on experience with LLM inference engines such as vLLM or SGLang, including serving large MoE models.
Strong GPU profiling and optimization skills covering memory, NCCL communication, CUDA graphs, and kernel bottlenecks.
Strong software engineering practice and ability to work fluently in large open-source codebases.
Proven ability to lead complex technical projects end to end, mentor engineers through code review and technical guidance, diagnose complex system issues, and collaborate with researchers and engineers.
Commitment to code quality, testing, and reproducibility.
Nice to have
Experience building or extending RL post-training frameworks such as NeMo-RL, verl, OpenRLHF, or TRL, especially asynchronous or non-colocated systems.
Experience training or serving models above 100 billion parameters with 100k+ token contexts.
Experience with Ray, Kubernetes, or Slurm on AWS, Azure, or Google Cloud.
Experience with numerical-parity testing between training and inference implementations.
Significant contributions to vLLM, Megatron, PyTorch, or similar open-source projects.
Publications or technical reports at ML or systems venues such as MLSys, NeurIPS, or OSDI.
Culture & Benefits
Flexible hybrid work model for office-based roles.
Work from anywhere for up to 8 weeks per year as part of flexible work arrangements.
Flexible vacation, two company-wide Mental Health Days, Headspace access, retirement savings, tuition reimbursement, incentive programs, and wellbeing resources.
Continuous learning, skills development, and career growth programs.
Two paid volunteer days annually and opportunities for pro-bono consulting and ESG initiatives.
Inclusive culture focused on customer impact, fast learning, collaboration, flexibility, and work-life balance.
Будьте осторожны
Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →