Post-Training AI Research Engineer for Drug Discovery

Helical; Greater London; On-site; GBP 110,000 - 150,000; Full time

Job summary

Helical is building in-silico labs for biology and transforming drug discovery with foundation models. The Applied Research Engineer - Post-Training will own the full post-training lifecycle, designing pipelines that adapt general models into therapeutic tools for pharma clients.

You’ll drive validation, experimentation, and end-to-end deployment, working with ML infrastructure and biology teams. You’ll decide what to fine-tune, how to validate biologically, and how to ship results to clients

Qualifications

• MSc or PhD in ML, computational biology or equivalent.
• Hands-on post-training techniques: fine-tuning, LoRA, DPO, RLHF.

Responsibilities

• Design and implement post-training pipelines that align biological foundation models to specific therapeutic contexts and client use cases.
• Build validation frameworks that connect model improvements to biological ground truth—using embeddings, perturbation data, and OpenTargets.
• Own experiments end-to-end: from hypothesis through training runs on distributed GPU infrastructure to analysis and client delivery.
• Collaborate with ML engineers on training infrastructure and with biologists to ensure outputs are scientifically meaningful.
• Contribute to our open-source tooling (helical-package) and shape the technical direction of our post-training capabilities as we scale.
• Stay at the frontier of post-training research and bring relevant advances into production.

Skills

Independent work

Communication

Experiment design

Education

MSc or PhD in Machine Learning or Computational Biology

Tools

Python

PyTorch

Job description

Helical is building in-silico labs for biology and transforming drug discovery with foundation models. The Applied Research Engineer - Post-Training will own the full post-training lifecycle, designing pipelines that adapt general models into therapeutic tools for pharma clients.

You’ll drive validation, experimentation, and end-to-end deployment, working with ML infrastructure and biology teams. You’ll decide what to fine-tune, how to validate biologically, and how to ship results to clients