Flagship Pioneering
ProFound Therapeutics: AI/ML Engineering - SAGE Platform Co-Op
Cambridge, MA USA +1 · Internship · Posted Oct 23, 2026
- Location
- Cambridge, MA USA
ProFound Therapeutics, Inc. - Experience
- No specific number of years stated; hands-on LLM experience may come from coursework, projects, or prior internships.
- Education
- Currently pursuing a B.S. or M.S. in Computer Science, Artificial Intelligence, Data Science, or a related field.
About the job
Build and improve SAGE, an internal multi-agent LLM platform that helps scientists analyze biological knowledge and accelerate drug discovery.
What you'll do
- Build and deploy agent tools, data integrations, document ingestion, and retrieval-augmented generation features.
- Develop SAGE chat collaboration features, evaluation and benchmarking pipelines, and automated quality checks.
- Monitor usage, performance, reliability, and LLM costs; write and document tested code and present results.
What they ask for
- Currently pursuing a B.S. or M.S. in Computer Science, Artificial Intelligence, Data Science, or a related field, and available for a full-time, 4–6 month co-op term.
- Strong Python skills.
- Hands-on LLM-building experience through coursework, projects, or prior internships (e.g., LLM APIs, RAG, agents, or MCP).
- Familiarity with Git and software engineering best practices, including testing, code review, and documentation.
Nice to have
- Experience with web or backend development, LLM evaluation or observability tools, knowledge graphs, vector databases, or scientific and biological data sources.
- Familiarity with cloud environments, containers, and CI/CD tooling; prior software or ML engineering internship or co-op experience; interest in biology or drug discovery.
Pay
$25 - $35 per hour (base pay)
Tools
Python, LLM APIs, RAG, MCP, FastAPI, React, TypeScript, Git
AI focus
Extend and operate a production multi-agent LLM platform for scientific research, including agent tools and integrations, retrieval over biological knowledge sources, answer-quality and hallucination evaluation, and monitoring.
Model-written summary · the original posting is authoritative