■ Company Overview Cheiron is a Stanford-founded AI company building an AI-native operating system for the pharma, biotech, and broader life sciences industry. Since launching in 2024, we've rapidly grown our global user base. In Korea, we reached 22% of industry users within six months and are now used across 200+ pharma and biotech companies as we continue to scale quickly. ■ About the Role We're looking for a Backend Engineer, Applied Agents to help build Cheiron's agent layer and the backend systems that power it. Cheiron's agents don't just generate answers. They carry out multi-step tasks inside real drug development workflows, call tools reliably, execute code in sandboxed environments, retain context across multi-step work, and deliver verifiable results to users. You'll design and ship the agent runtime, harness, memory, and tool execution layers that make all of this possible. We build products that run in real customer environments — not research demos or prototypes. We're looking for a hands-on builder who can quickly structure ambiguous problems and turn them into highly polished products. You don't need to be an AI researcher or a life sciences domain expert, but we care deeply about strong engineering fundamentals and real experience designing, building, and operating LLM systems. ■ What You'll Do - Design, build, and operate Cheiron's agent layer: agent runtime/harness, tool-calling infrastructure, sandboxed code execution, agent memory, and the reliability and observability of multi-step tasks - Design and build backend services and APIs on Python, FastAPI, and Postgres - Design and operate vector DB, semantic search, and RAG systems, and tune search quality and latency - Build large-scale ingestion, cleaning, and indexing pipelines for life sciences data, including academic papers, clinical, regulatory, safety, and patent sources - Build systems that ground AI-generated outputs in source data with traceable, verifiable citations so they can be trusted in real pharma and biotech work - Work directly with founders, domain experts, and early customers, owning outcomes rather than just closing tickets ■ Requirements - 2+ years of experience shipping and operating production backend systems end to end - Strong backend engineering fundamentals: a deep understanding of API design, data modeling, databases, and system reliability - Hands-on production experience with AI-driven systems such as LLMs, RAG, and vector DBs - Practical understanding of how agent systems actually work: you've built — or at least deeply traced — agent runtimes, harnesses, tool calling, sandboxing, and memory - The drive to set your own priorities and ship quickly, even when specs are incomplete - A balance of fast shipping velocity and sound engineering judgment - Active use of AI coding tools such as Claude Code and Cursor ■ Nice to Have - (Most important) Experience building an in-house agentic harness: if you've designed and built internal tools or harnesses to boost your team's agentic development productivity, that's the strongest signal for us - Hands-on experience designing and operating LangGraph, agent frameworks, or RAG systems in production - Experience building enterprise SaaS, or pharma, biotech, or healthcare products - Familiarity with life sciences domain data such as academic literature, clinical trials, regulatory documents, and patents - Experience deploying and operating production systems on AWS, Terraform, and Kubernetes, with a focus on reliability and observability - Experience at a seed or early-stage startup ■ Benefits & Perks - 401(K) retirement plan - Health insurances (Medical/Dental/Vision) - Meal allowance (Lunch, Dinner) - Transportation support for early starts and late nights - In-office snack bar and additional commuting and work travel support ■ Interview Process Screening ＞ Take-home Assignment ＞ Technical Interview ＞ Cultural Interview