5 дней назад
Machine Learning Research Engineer (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Machine Learning Research Engineer (AI): Building evaluation suites, training-data pipelines, and specialized language models for an autonomous research engine with an accent on post-training, continual learning, and scientific workflow automation. Focus on designing task-specific evaluations, solving regression and forgetting across repeated training cycles, and improving model quality, cost, and latency.
Location: In-office in Barcelona, Spain; candidates should be ready to work from Barcelona or move there.
Company
develops Nextflow and the Platform to help scientists and organizations scale complex data workflows.
What you will do
- Design and build evaluation suites for an autonomous research agent.
- Turn deployment run data into training-ready corpora.
- Post-train small open-weight models using supervised fine-tuning, preference tuning, and adapters.
- Run repeated training cycles on evolving data while managing regression and forgetting.
- Collaborate with the Founding Scientist and application engineers on the Next-Gen roadmap.
- Participate in journal clubs and apply current research in specialized model post-training.
Requirements
- Hands-on experience post-training open-weight language models with SFT and at least one preference-tuning method.
- Experience retraining models on changing data and applying replay, data mixing, adapter strategies, and evaluation gating.
- Experience building task-specific evaluation harnesses and defining metrics.
- Fluent Python skills and experience with PyTorch, Transformers, TRL, PEFT, vLLM, or equivalent tools.
- Substantive open-source contributions or fine-tuned models published on the Hugging Face Hub.
- Willingness to work in-office in Barcelona or move there.
Nice to have
- Experience in computational biology, chemistry, or pharma.
- Familiarity with Nextflow or workflow orchestration.
- Experience evaluating agentic, multi-step, tool-using systems.
- Experience with distillation or small-model specialization from frontier models.
- Publications or technical reports in post-training or continual learning.
Culture & Benefits
- Flexible working hours and an international environment with more than 25 nationalities.
- Continuous skills development, a learning and development budget of 1,000 USD per year, and a home office allowance valued at over 1,000 USD.
- 23 vacation days per year, three additional company days in December, and public holidays according to the employee's location.
- Equity, private health insurance, private life insurance, and access to the Oliva mental health app.
- Team retreats, bonding activities, and a fast-growing startup environment.
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 →