What are the responsibilities and job description for the AI Agent Builder position at Harrison Clarke?
AI Agent Engineer | Palo Alto | Seed-Stage, VC-Backed
We are working with a well-funded, early-stage AI startup based in Palo Alto that is building next-generation enterprise AI agents capable of automating complex, multi-step business workflows.
This is not a chatbot role. The team is focused on designing reliable, production-grade agent systems that can reason, retrieve context, call tools, and execute tasks safely across enterprise platforms such as SAP, Salesforce, and Workday.
Backed by top-tier institutional investors and led by founders with deep AI research and enterprise systems experience, the company is assembling a small, high-caliber engineering team to define what scalable, real-world agent architecture looks like.
The Role
As an AI Agent Engineer, you will:
- Design and build multi-step, tool-using AI agents
- Architect RAG pipelines and memory systems
- Implement planning, execution, and verification loops
- Develop guardrails and reliability mechanisms for enterprise environments
- Optimize inference performance and latency
- Deploy production systems across modern cloud infrastructure
This is a hands-on engineering role with real ownership. Early hires will shape the core agent framework and system design.
What They're Looking For
- Strong experience building LLM-powered systems in production
- Experience with RAG, embeddings, and vector databases
- Familiarity with agent frameworks such as LangGraph, LangChain, AutoGen, or similar
- Experience with tool-calling, API orchestration, and multi-step reasoning workflows
- Strong Python skills and cloud deployment experience
- Comfort operating in small, fast-moving teams
Ideal candidates have built more than just LLM demos. We are looking for engineers who understand how agents fail, how to design around those failures, and how to ship reliable systems.
Why This Opportunity
- Seed-stage but institutionally backed
- Lean, engineering-first culture
- Real ownership and architectural influence
- Focused on meaningful enterprise automation problems
- Bay Area based
If you are interested in building real AI agents that execute work rather than generate text, I would be happy to share more details confidentially.