Job Details

Title: Applied AI Engineer

Location: San Francisco, CA (On-site, 4–5 Days/Week)

Compensation: $160,000 – $280,000 Base + Equity

Work Authorization: Must be Authorized to Work in the US without Restriction

The Opportunity

We're working with a well-funded, early-stage AI startup redefining how scientific research is conducted. Backed by leading AI investors and operating at the cutting edge of autonomous AI agents, this company is building technology that compresses months of R&D into a single day and is already seeing serious inbound from some of the world's largest life sciences organizations.

This is a rare opportunity to join as a founding-tier AI engineer, own the development of production agents used by world-class researchers, and help define what the next generation of AI-driven scientific discovery looks like.

The Role

As an Applied AI Engineer, you'll own the development and maintenance of production AI agents on the platform. This is a builder role, you'll architect new agent capabilities, prototype with internal science and engineering teams, and ship agents that researchers depend on daily.

Expect to spend 70% building (agent architectures, prompting strategies, tool integrations, evaluation frameworks) and 30% in the field with R&D partners, understanding real scientific workflows and validating what you've built. You'll also help shape product direction by translating field insights into prioritized roadmaps.

What You'll Own

• Architect and ship production AI agents used by pharma, biotech, and academic research teams daily
• Build and iterate on prompting strategies, context engineering, and tool-use pipelines
• Design and implement evaluation and benchmarking frameworks for agent systems
• Spend meaningful time with R&D partners (including VP and C-suite at leading pharma companies) to understand workflows and validate builds
• Translate field insights and internal research into product roadmap direction
• Partner cross-functionally with science, engineering, and product teams

What We're Looking For

Must-Haves:

• 3+ years professional software engineering experience shipping production systems real users depend on
• Hands-on LLM/AI application experience: prompting, context engineering, agent architectures, evaluation frameworks (not just theory)
• Agentic AI experience: LangChain, RAG, agent orchestration, tool-use pipelines (non-negotiable)
• Strong Python and/or TypeScript proficiency
• Builder mentality: comfortable working from scratch, driving clarity in ambiguous situations, wearing multiple hats
• Ability to work across engineering, science, and product, you can translate between technical and domain teams

Nice-to-Haves:

• Life sciences, biomedical research, scientific computing, or technical R&D workflow experience
• Experience building evaluation or benchmarking frameworks for AI systems
• Open-source contributions to AI or scientific tooling
• Science background (biology, chemistry, physics, computational science)

The Stack

Python · TypeScript · LangChain · RAG · Agent Orchestration Frameworks · LLM Tooling · Prompt & Context Engineering · Evaluation Frameworks · Tool-Use Pipelines · Benchmarking Systems

Why This Role

• Ownership: End-to-end ownership of some of the most powerful AI agents in life sciences, from prototype through production
• Impact: Your agents will be used daily by researchers at leading global pharma and biotech organizations
• Visibility: Work directly with R&D partners including VPs and C-suite at top pharma companies with 30% field time, not just engineering in isolation
• Product Influence: Translate what you learn in the field directly into roadmap decisions, your voice shapes what gets built next
• Speed: Small, elite team scaling fast, decisions you make now will define the platform for years
• Comp: $160,000 – $280,000 base + meaningful early-stage equity.
• Mission: Build the AI agents that let scientists run months of research in a single day

What This Is Not

This is not a research role, a fine-tuning role, or a role for engineers who've only experimented with LLMs in side projects. This is for someone who has shipped production agentic AI systems that real users depend on, and who wants to do that at a company where AI is the entire mission, not a feature.

Interested or know someone who might be? Apply below or reach out directly.