SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Healx is an AI-powered tech bio company redesigning drug discovery for rare diseases. With 10,000 rare diseases affecting 400 million people globally and 90% lacking approved treatments, Healx combines data, AI, and deep pharmacology expertise to develop treatments faster and more cost-effectively than traditional drug discovery.
You will join as an Agentic AI Engineer reporting to the Director of Tech Strategy. Your core mission is to design and build LLM-agentic workflows that power Healx's drug discovery platform. Working closely with scientists across the company, you'll translate their most pressing drug discovery questions into agent designs, then build and refine agents that reason over Healx's knowledge graph, proprietary methods, scientific literature, and other sources to generate, triage, and rationalize testable therapeutic hypotheses for rare diseases.
You won't start from scratch. Healx has already adopted an agentic framework (Python, ADK, AgentSpace) with workflows delivering impact across the platform. A significant part of your role is helping adapt as this fast-moving field evolves. You'll work alongside the technical lead owning agentic infrastructure, with scope to take ownership of specific workflows and grow your influence.
Key Responsibilities:
- Build and refine AI-agentic workflows on the internal framework to help drug discovery scientists generate the best testable therapeutic hypotheses.
- Translate discovery problems into agent designs, determining where agentic approaches genuinely add value.
- Integrate agents with the knowledge graph, proprietary methods, scientific literature, and other sources so they reason over the right evidence.
- Expand and maintain existing GenAI tools to keep pace with team needs and the fast-moving ecosystem.
- Contribute to evaluation strategies for agentic workflows, helping shape sensible evals, testing, and quality standards as the practice matures.
- Write clear, maintainable, well-documented code that others can build on.
Success Metrics (3 months): Master the agentic framework and discovery problems it serves; make first contributions to the codebase; build working relationships with scientists and engineers; turn feedback into concrete improvements.
Success Metrics (6 months): Take at least one agentic workflow from idea to production tool used in the drug discovery pipeline; operate with real independence, owning workflows end-to-end and proposing improvements.
Healx operates from a modern, accessible office in central Cambridge (near the train station) with a hybrid, highly collaborative model that values synchronization and pair programming. The team includes colleagues with decades of accumulated expertise in their domains, committed to supporting your skill development and career growth.
Requirements:
- Built and shipped LLM-agentic systems that deliver real value in production (agents using tools, orchestrating multi-step workflows, behaving reliably).
- Minimum 2 years of software engineering or ML engineering experience.
- Strong software engineering fundamentals: write clear, tested, maintainable Python code that others can build on.
- Fluency with modern LLM/agent toolkit: model APIs, prompting, tool use, RAG, and emerging patterns (MCP, agent frameworks, evals).
- Comfortable working closely with non-engineers; ability to sit with scientists to understand their actual needs.
Bonus Qualifications:
- Experience in drug discovery, biology, or other life science domains.
- Familiarity with knowledge graphs or reasoning over structured/heterogeneous data.
- Experience building evaluation harnesses or testing strategies for LLM systems.
- Track record of picking up unfamiliar domains quickly and becoming useful fast.
- Interest in or experience with biotech/techbio and its impact on patient outcomes.
