> Markdown version of [/jobs/ext/4690371-founding-ai-agent-engineer](https://www.wearedevelopers.com/jobs/ext/4690371-founding-ai-agent-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Founding AI Agent Engineer

- **Company:** BRAUN, GEORGE CONSULTING
- **Location:** San Francisco, CA, United States
- **Contract:** Permanent contract
- **Skills:** LangGraph Framework, Artificial Intelligence, Automation of Tests, Next.js, Software Safety, Short Message Service, Software Engineering, Systems Integration, Core Voice Platform, Large Language Models, Multi-Agent Systems, Prompt Engineering, Agentic-AI, Virtual Agents
- **Published:** November 12, 2026
- **Apply:** https://www.careerbuilder.com/job-details/founding-ai-agent-engineer-san-francisco-ca--4d1ea6d9-5b0b-468a-8cff-8beccc566ad4

## About the Role

* 0-4 years in software engineering, agent engineering, or applied AI.
 * Has launched conversational AI agents to real consumers at scale, not just internal tools or side projects.
 * Strong working knowledge of LLM testing, live-system tracing, prompt chaining, and agent control loops.
 * Able to work on-site 6 days a week in the San Francisco Bay Area.
 * Open to candidates on US work visas (OPT, H1B transfers).

Skills Required

 * Conversational AI & multi-agent systems
 * LLMs, prompt engineering & evals
 * Agent tracing & observability
 * LangGraph, Vercel AI SDK
 * Voice AI & SMS/email integrations
 * CRM / clinical trial system integrations
 * AI safety guardrails

## Description

* Design and launch a group of AI agents that work together to recruit and support patients across many active clinical trials.
 * Own the full improvement cycle for the agents: review live conversations, turn failures into automated tests, and keep tuning how the agents run.
 * Connect health-profile data so agents reach patients at the right moment and quickly line up large numbers of pre-screened candidates when a trial opens.
 * Extend the agents across voice, text messages, and email while keeping conversations consistent, personal, and empathetic.
 * Bring in new agent frameworks, memory approaches, and tools to strengthen the core product.
 * Make sure the agents are reliable and safe in a heavily regulated healthcare setting.