Software Engineer, AI Agents
Foresite Labs
$200,000 - $240,000 / year
Need a reasonable accommodation to apply or interview? Contact us.
JobMinglr uses automated technology to recommend jobs based on profile information and job preferences. Match Score does not determine eligibility for a position, prevent a user from viewing or applying to a job, or make hiring decisions on behalf of an employer.
Description
About Foresite Labs
Foresite Labs is a translational R&D team that derives insights from precision measurement and population-scale biology and genetics to address unmet clinical needs. We use human genetics to systematically dissect and understand human disease biology and develop and critically evaluate therapeutic hypotheses. We engage in translational research, transforming basic insights into therapeutic opportunities. Our work supports drug discovery and company formation, and provides the core around which new ideas are realized and incubated. We offer competitive salaries, excellent benefits, a flexible work environment, and the opportunity to learn from top thinkers in various disciplines. Foresite Labs is headquartered in San Francisco.
The Role
We're hiring a software engineer to help create the data pipelines, machine learning infrastructure, and agentic systems that power our science. On the agentic side, that means writing the code that runs the loop an AI agent operates inside, not one-off prompts — designing how agents plan, call tools, orchestrate sub-agents, keep memory, check their own work, and decide when they're done, then shipping them into production to do real work across our genetics and drug-discovery pipeline. This is "loop engineering": the model writes the prompts; you build and harden the machinery around it.
It's a full-stack, high-ownership role on a small senior team. You'll take systems from a blank repo to production and work shoulder-to-shoulder with ML engineers, computational biologists, and scientists.
What You’ll Do
- Build agent loops end-to-end — tool/function calling, planning, sub-agent orchestration, memory, and control flow that runs reliably and unsupervised
- Build the eval and observability harness — the tests, metrics, tracing, and guardrails that make agentic systems trustworthy in production
- Turn scientific and operational workflows into agents that take real action across our platform
- Ship full-stack features: React/Tailwind front ends and backend services that put these agents in front of scientists
- Design and launch production-grade systems, services, data pipelines, and machine learning infrastructure from scratch
- Use frontier AI-native development tools (e.g. Claude Code) to move fast
- Uphold engineering standards through code review, testing, and peer mentorship
What You’ll Bring
- 5+ years of professional software engineering, with greenfield systems you've shipped to production
- Hands-on experience building LLM/agent systems in production — agent loops, tool calling, RAG, and evals (frameworks like LangChain, Pydantic AI, or equivalent)
- Fluency in TypeScript plus a modern backend language (Python, Go, or Java)
- Working knowledge of AWS/GCP/Azure core services and infrastructure-as-code (Terraform, CDK, or Pulumi)
- Comfort in Unix, Docker, and CI/CD pipelines
- Early-stage execution mindset and strong ownership
- Clear communication across technical and non-technical audiences
Nice to Have
- Experience with LLM eval / observability tooling (e.g. LangSmith) and prompt/loop iteration at scale
- Multi-agent orchestration in production
- Interest in biology, genetics, or human health — no life-sciences background required
Why Foresite Labs
- Build frontier agentic products that measurably accelerate real drug discovery
- Greenfield work, end-to-end ownership, and architecture decisions without bureaucracy
- A small, senior team working directly across software, ML, and biology
- Competitive base of $200,000–$240,000, meaningful equity, comprehensive healthcare, generous PTO, and hybrid/remote flexibility
Location: San Francisco, CA
Salary range: $200,000 - $240,000
Foresite Labs is an equal opportunity employer. We thrive on diversity and collaboration.
About this role
Foresite Labs is building AI agent systems to accelerate drug discovery, and they're looking for a software engineer to own the full stack—from agent loop architecture to production deployment. This role sits at the intersection of infrastructure and applied AI: you'll design how agents plan, call tools, manage memory, and orchestrate sub-agents, then ship those systems into a real genetics and drug-discovery pipeline. It's not prompt engineering; it's the machinery that makes agentic workflows reliable and unsupervised at scale.
You'll need at least five years of production software engineering experience, with concrete hands-on work building LLM and agent systems—tool calling, RAG, evals, frameworks like LangChain or Pydantic AI. TypeScript fluency plus Python, Go, or Java; AWS/GCP/Azure; Docker and CI/CD are table stakes. The role is full-stack: you'll build eval and observability harnesses, ship React/Tailwind frontends alongside backend services, and design data pipelines and ML infrastructure from scratch. You'll work on a small senior team alongside ML engineers, computational biologists, and scientists, with genuine greenfield ownership and no bureaucratic overhead.
This is in-office in San Francisco with a base of $200,000–$240,000 plus equity. The role suits someone who thrives on end-to-end ownership, moves fast with frontier AI tools, and can communicate clearly across technical and non-technical teams. Biology background is not required, but genuine interest in how your work impacts drug discovery and human health would fit the culture.
This role's local market on JobMinglr
Pay for this role
$200,000 to $240,000 per year
That is the employer's posted range, and where you land in it is usually decided in one conversation. Before that conversation, run the numbers: analyze an offer for this role.
Before you apply, worth reading
How JobMinglr reads this job
Every listing here is scored against your profile before you apply: skills overlap, experience level, location and work arrangement, each weighted and explained. You see the score and the reasons, not just a list. How the matching works.