What are the responsibilities and job description for the MLE - Agent Builder position at Harrison Clarke?
We're partnering with a Series A AI startup building autonomous agents that don't just answer questions. They reason, plan, execute, recover from failure, and improve over time.
The company is still small enough that every engineer shapes the product, architecture, and technical direction. You'll work directly with the founders to build the systems that define the company over the next decade.
This isn't a feature engineering role.
This is your opportunity to build an AI platform from the ground up.
The Opportunity
- We're looking for a Founding AI Engineer to own the complete lifecycle of production AI agents.
- You'll take ideas from zero to production, designing the architecture, building the agents, shipping them to customers, measuring performance, and continuously improving them through real-world feedback.
- You'll own the entire stack, from orchestration and reasoning through infrastructure, evaluation, deployment, and iteration.
- If you enjoy building products from scratch and want genuine ownership, this is exactly that.
What You'll Do
- Build production AI agents from 0→1
- Design multi-agent systems capable of planning, reasoning and executing complex workflows
- Build orchestration frameworks, tool use, memory and retrieval systems
- Develop evaluation pipelines that measure agent quality in production
- Create feedback loops that continuously improve agent performance
- Own prompt engineering, context management and reasoning strategies
- Build scalable backend services that power AI products
- Ship features directly to customers and iterate rapidly from usage
- Make fundamental architecture decisions alongside the founding team
- Help build the engineering culture from the earliest days
What They're Looking For
- Experience building and shipping production AI applications
- Strong Python engineering skills
- Experience with LLM APIs and modern agent frameworks
- Built products using tool calling, RAG, structured outputs or multi-step reasoning
- Comfortable owning backend systems, APIs and infrastructure
- Strong product instincts and ability to operate with ambiguity
- Someone who thrives in fast-moving startup environments
- Excited by true ownership rather than working on a small piece of a large organisation
Nice to Have
- LangGraph, OpenAI Agents SDK, CrewAI or similar frameworks
- Experience building multi-agent architectures
- LLM evaluation or observability experience
- Reinforcement learning or online learning systems
- Distributed systems or platform engineering experience
- Kubernetes, Docker or cloud infrastructure
- Experience founding or joining an early-stage startup