Applied Scientist, Post-training
Core
Developing post-training pipelines and frameworks to enable large multimodal models to reason, explain logic, and make verifiable decisions for high-stakes industries like medtech and aerospace.
Role type
Applied Scientist (Post-training & Alignment)
Builds
Domain-specific reasoning systems and agentic AI for medtech, aerospace, and advanced manufacturing.
Domain
AI/ML, Large Language Models, Reasoning Systems
Deliverable
production ML models
Required skills
Transformer-based model training, RLHF, DPO, reward modelling, Python, PyTorch
Preferred skills
Multimodal reasoning, evaluation and verification, interpretability research
Technologies
PyTorch, LLMs, VLMs, MLLMs
Responsibilities
Develop and optimise post-training pipelines, implement reward modelling for reasoning depth and factual accuracy, build evaluation frameworks for verifiable human-aligned behaviour, run end-to-end experiments with proprietary and synthetic datasets, deploy methods into production.
Seniority
Mid-Senior, hands-on IC
