{"id":1952214,"url":"https://alion.io/job/jcat-ml-engineer-middlesenior","title":"ML Engineer (Middle+/Senior)","company":{"id":21126,"name":"JCat","domain":"jcat.ru","url":"https://alion.io/company/jcat","size_band":"11-50","is_staffing_agency":true,"employer_type":"agency","is_intermediary":true,"listed_via":null,"ats_vendor":"HeadHunter","truth_index":{"grade":"D","score":52,"open_postings":39,"ghost_share":0.692,"stale_share":0,"repost_share":0.769,"time_to_fill_p50_days":3,"computed_at":"2026-10-09T06:01:00Z"}},"role":"AI/ML","role_family":"AI/ML","seniority":"senior","employment_type":null,"work_mode":"on_site","remote_scope":null,"remote_scope_basis":null,"remote_working_hours":null,"hiring_geo_confidence":"structured","locations":["Moscow, Russia"],"countries":["RU"],"hiring_countries":[],"hiring_countries_total":0,"salary":{"min":220000,"max":260000,"currency":"RUB","period":"month","gross":null,"usd_annual":36432},"salary_estimate":null,"experience_years_min":2,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"AI Agents","optional":false},{"name":"ComfyUI","optional":false},{"name":"CUDA","optional":false},{"name":"CUDA Toolkit","optional":false},{"name":"Diffusers","optional":false},{"name":"Docker","optional":false},{"name":"Hugging Face","optional":false},{"name":"Llama","optional":false},{"name":"LLM","optional":false},{"name":"LoRA","optional":false},{"name":"NLP","optional":false},{"name":"PEFT","optional":false},{"name":"QLoRA","optional":false},{"name":"Qwen","optional":false},{"name":"RAG","optional":false},{"name":"SDXL","optional":false},{"name":"Sentence-Transformers","optional":false},{"name":"TRL","optional":false},{"name":"Unsloth","optional":false},{"name":"vLLM","optional":false},{"name":"FastAPI","optional":true},{"name":"Python","optional":true},{"name":"Stable Diffusion","optional":true},{"name":"Transformers","optional":true}],"status":"live","first_seen_at":"2026-10-06T10:25:05Z","employer_posted_date":"2026-10-06","last_verified_at":"2026-10-09T00:12:41Z","board_verified":true,"closed_at":null,"days_open":3,"trust":{"level":"ok","repost_count":null,"flags":[],"days_open":3},"description":"Вакансия компании: ООО \"КОМБЛОК\"\nMiddle AI / LLM Engineer\nИщем прикладного AI-инженера для запуска и адаптации open-source моделей под наши продукты.\nНаш фокус - не теоретический R&D, а быстрая доставка готовых ML-решений в продакшен: взять проверенную открытую модель, дообучить под наши сценарии и задеплоить на собственном GPU-сервере.\nПродуктовый бэкенд, базы данных и сбор данных закрывает основная команда разработки. Ваша зона ответственности - подбор и тюнинг моделей, инференс, производительность на GPU и качество генерации.\nЧем предстоит заниматься:\n Сервинг и инференс: развернуть и сопровождать open-source модели (Qwen, Llama и др.) на выделенном сервере с GPU Nvidia через vLLM.\n Стилизация и адаптация (LoRA): дообучать модели под определенную персону (сленг, подача, характерные мысли) через LoRA / QLoRA с сохранением связности и логики ответов.\n Семантический поиск и классификация (RAG): собирать пайплайны контекстного поиска на базе эмбеддеров для русского языка (ruBERT / BGE / Sentence Transformers).\nAI-агенты: настраивать логику вызова инструментов и промпт-пайплайны под автономные задачи.\n Генерация изображений: поднять и настроить готовый сервис генерации картинок на базе open-source моделей (Flux / SDXL).\nЧто для этого нужно:\nКоммерческий опыт в прикладном ML / NLP от 2 лет.\nОпыт запуска и инференса LLM на видеокартах Nvidia (понимание vLLM, Docker, работы с CUDA и контролем видеопамяти без падений в OOM).\nПрактический опыт файн-тюнинга через LoRA / QLoRA (Hugging Face PEFT, TRL, Unsloth).\nПонимание подготовки данных: фильтрация шума, сбор пар «промпт-ответ», предотвращение выгорания логики модели (catastrophic forgetting).\nБазовый опыт (или четкое понимание), как поднять готовый инференс для генерации картинок (Diffusers, ComfyUI API).\nБудет плюсом:\n Уверенный прикладной Python: умение написать чистый скрипт предобработки данных, обернуть инференс в простую ручку на FastAPI или автоматизировать пайплайн без спагетти-кода.\nОпыт работы с легковесными NLP-моделями (ruBERT-tiny2) под задачи классификации текстов.\nУсловия:\nФормат: первые 1-2 месяца - офис в Москве, далее - удобный гибрид по запросу.\nИнфраструктура: выделенный сервер с мощными GPU Nvidia под задачи без очередей и квот.\nКоманда и процессы: быстрый цикл принятия решений, понятные прикладные задачи и готовая бэкенд-команда на поддержке.\nДополнительные преимущества:\nОбучение и тренинги","description_format":"text","description_chars":2398,"description_truncated":false,"requirements":{"experience_years_min":2,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":null,"security_clearance":false,"languages":[{"language":"Russian","level":"All levels","optional":false}]},"benefits":[],"hiring_locations":[],"hiring_excludes":[],"relocation_offered":false,"industries":[],"lifecycle":[{"event":"open","at":"2026-10-06T10:28:22Z"}],"visa":[],"liveness":{"score":23,"band":"cold","label":"Long shot","p_open":1,"p_active":0.293,"p_room":0.788,"age_days":2,"expected_fill_days":3,"reasons":["conf:5","agency","stale_co","velocity","win:late"],"computed_at":"2026-10-09T06:01:00Z"},"pay":{"stated_usd_annual":36432,"is_top_pay":false},"html_url":"https://alion.io/job/jcat-ml-engineer-middlesenior","json_url":"https://alion.io/job/jcat-ml-engineer-middlesenior.json","meta":{"generated_at":"2026-10-10T00:49:11Z","cache_seconds":300,"methodology":"https://alion.io/methodology","terms":"https://alion.io/terms","contact":"https://alion.io/contact","api":"https://alion.io/developers","about":"Alion is a live layer of people, companies and AI agents: who they are, whether they are real and active right now, what they do and how to work with them, readable by people and by agents and paid per call.","catalog":"https://alion.io/catalog.json","usage":{"tier":"crawler","counted_by":"address","units_charged":1,"used_today":1300,"day_limit":5000,"remaining_today":3700,"minute_limit":60,"resets_at":"2026-10-11T00:00:00Z"}},"offers":[{"id":"company.slices","title":"One company in depth, by slice","status":"live","price":{"credits":0.02,"usd":0.002,"plus_per_slice":{"credits":0.05,"usd":0.005}},"unit":"per company, plus each slice with data","note":"the employer in depth","call":{"mcp_tool":"get_company","arguments":{"id":21126},"rest":"https://alion.io/mcp/rest/get_company?id=21126"},"human":"https://alion.io/catalog?offer=company.slices&for=job%2Fjcat-ml-engineer-middlesenior"},{"id":"market.stats","title":"A market slice: pay, demand and time to fill","status":"live","price":{"credits":1,"usd":0.1},"unit":"per slice","note":"pay, demand and time to fill for this role and place","call":{"mcp_tool":"market_stats"},"human":"https://alion.io/catalog?offer=market.stats&for=job%2Fjcat-ml-engineer-middlesenior"},{"id":"job.search","title":"Open jobs by role, technology, place, pay and visa","status":"live","price":{"credits":0.02,"usd":0.002},"unit":"per posting in a list","note":"similar open postings","call":{"mcp_tool":"search_jobs"},"human":"https://alion.io/catalog?offer=job.search&for=job%2Fjcat-ml-engineer-middlesenior"},{"id":"company.verify","title":"Is this company real and active right now","status":"pilot","price":null,"unit":"per company","request":{"url":"https://alion.io/catalog/request","method":"POST","body":"{\"offer\": \"company.verify\", \"for\": \"job/jcat-ml-engineer-middlesenior\", \"note\": \"what you need it for\"}"},"human":"https://alion.io/catalog?offer=company.verify&for=job%2Fjcat-ml-engineer-middlesenior"}]}