Software Developer, Agentic AI — Intelligent Agents and Retrieval Systems for Computational Drug Discovery

This is a rare chance to bridge the gap between cutting-edge agentic AI and atomic-resolution drug discovery at an elite research organization. You will not just be throwing wrappers around LLMs; you are building production-scale retrieval systems and intelligent agents that actively accelerate life-saving therapeutics.

What You'll Actually Be Doing

You will spend your days architecting systems that connect modern generative AI directly to heavy-duty scientific software and machine learning pipelines. Expect to prototype multi-step agentic workflows, wrangle complex data pipelines, and figure out how to make AI models learn smarter and faster. The challenge here is less about writing boilerplate web apps and more about understanding messy scientific data and building robust tools that real researchers will actually rely on every day.

The Core Tech Stack

Python is your primary weapon here, coupled with deep familiarity with Retrieval-Augmented Generation (RAG) architectures and autonomous agent loops. You need to know how to build reliable infrastructure that doesn't just work in a Jupyter notebook, but scales cleanly to production. Experience with workflow automation and code generation tools will also be critical since you are designing systems that orchestrate complex, multi-step scientific processes.

Interview Expectations

Expect the engineering panel to press you hard on how you handle state management, error recovery, and hallucination mitigation in complex agentic loops. They want to see how you reason about failure modes when an LLM drops the ball mid-workflow. Another major focus will be system architecture, specifically how you design low-latency RAG systems that can ingest, index, and retrieve massive volumes of specialized scientific data without choking.

Application Advice

Skip the generic web dev fluff on your resume and lean heavily into systems architecture, backend engineering, and any prior exposure to AI pipelines or scientific tooling. If you have built custom workflow automation frameworks, distributed Python systems, or complex RAG setups, put those right at the top. Make sure keywords like 'Agentic AI', 'RAG', 'Python', 'Workflow Automation', and 'System Architecture' are front and center to clear the ATS and catch the eye of engineers who value raw technical versatility.