Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
$208к-312к/год
$190к$220кмедиана$280к
По похожим ролям (50) ~ $220к/год
На СНГ рынке: $4.5к/мес ($3.7к - $6к)
91
Крутая вакансия
развернуть
Чётко очерченная и технически амбициозная роль в компании AI-инфраструктуры с ответственностью за путь модели до production; главный компромисс - узкая специализация на собственных ускорителях.
Кликните для подробной информации
Узкая специализацияСовременный AI-стекОтветственность за результатСфера AI-инфраструктуры
Оценка от Hirify AI
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Senior Principal Model Bring-up Engineer (AI): Bringing open-weight LLM architectures from Hugging Face to production serving on SambaNova RDU dataflow accelerators with an accent on model adaptation, multi-chip mapping, compilation, and strict numerical validation. Focus on building shared inference modules for continuous batching, speculative decoding, FP8/FP4, tensor and expert parallelism, and debugging across models, compilers, and hardware.
Location: Austin, Texas, United States, or San Jose, California, United States
Base salary: $208,000–$312,000 USD per year
Company
hirify.global develops AI infrastructure and a full-stack inference platform based on RDU dataflow accelerators and SambaRack rack-scale hardware.
What you will do
Bring up dense, mixture-of-experts, and hybrid Mamba/attention LLMs from model adaptation through production serving.
Convert checkpoints, map models across chips and hosts, compile them, and validate numerical correctness against reference implementations.
Build shared inference modules for continuous batching, paged KV caching, prompt caching, chunked and incremental prefill, dynamic tensor shapes, speculative decoding, FP8/FP4, and tensor/expert parallelism.
Debug issues across models, compilers, and hardware, landing fixes in reusable shared components.
Own a model family from first compile to served and merged, while shortening a long-running team workflow.
Requirements
Bachelor’s degree in Computer Science, Electrical Engineering, or a related field.
8+ years of industry experience in ML systems, inference, or compiler engineering.
Strong Python and PyTorch internals experience, plus some C++.
Deep transformer-inference knowledge covering KV caching, GQA, sliding-window attention, MoE routing, prefill versus decode, batching, and speculative decoding.
Hands-on experience with accelerator bring-up, ML compilers such as MLIR, XLA, TVM, Triton, or torch-mlir, or inference engines such as vLLM, SGLang, or TensorRT-LLM.
Rigor in numerical debugging and comfort with Bazel, Slurm, and large monorepos.
Clear writing and evidence-backed claims.
Nice to have
Master’s or Ph.D. in Computer Science, Electrical Engineering, or a related field.
Experience with dataflow or spatial compilers, kernel DSLs, FP8/FP4 inference, Mamba and SSMs, expert parallelism, multi-agent AI engineering workflows, or open-source inference/compiler contributions.
Culture & Benefits
Full-time US employment with base salary, equity, and benefits.
Medical insurance with 95% employee premium coverage and 77% dependent premium coverage, plus employer-funded HSA contributions.
Dental, vision, disability, life, AD&D, and flexible spending account options.
Well-being benefits including Headspace, Gympass+, One Medical, and counseling through an Employee Assistance Program.
Equal Opportunity/Affirmative Action employer.
Hiring process
Submit a separate application form for each position for which you wish to be considered.
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 →