Research Engineer - Post-Training

About the role

Helical is seeking an Applied Research Engineer to lead the post‑training lifecycle of biological foundation models, turning general‑purpose AI into therapeutic‑specific tools for pharma clients.

Key responsibilities

• Design and implement post‑training pipelines that align foundation models to therapeutic contexts and client use cases.
• Build validation frameworks linking model improvements to biological ground truth using embeddings, perturbation data, and external resources.
• Own end‑to‑end experiments from hypothesis through distributed GPU training, analysis, and client delivery.
• Collaborate with ML infrastructure engineers and biologists to ensure scientific relevance and production readiness.
• Contribute to open‑source tooling and guide technical direction of post‑training capabilities.

Required profile

• MSc or PhD in Machine Learning, Computational Biology, or related field, or equivalent industry experience.
• Hands‑on experience with post‑training techniques such as SFT, LoRA, DPO, or RLHF.
• Strong ability to work autonomously, make decisions with incomplete information, and communicate technical trade‑offs across teams.

Required skills

• Python
• PyTorch
• Transformer architectures
• Distributed GPU training (multi‑GPU, multi‑node, NCCL, DeepSpeed, FSDP)
• Post‑training methods (SFT, LoRA, DPO, RLHF)