VALE Partners have been engaged by an early-stage startup to hire their founding engineer. They're building a consumer AI product heading into beta, and are past the question of whether it should exist and focused entirely on getting it into users' hands. You'll have a direct line to the founder and meaningful input into the technical direction of the product.
Role Overview
- Shape the agent layer the product runs on, covering orchestration, tool use, and how agents reason and hand work to each other
- Teach the agents to ingest content accurately, working across VLMs, audio models, and retrieval that spans both
- Stand up the backend that carries a growing user base, from queueing and processing through to how workloads run on Kubernetes
- Run the model lifecycle end to end across pipelines, deployment, observability, and cost
What You'll Ideally Have
- A track record of running agentic AI in production rather than in prototypes, whether that is an LLM application, an ML pipeline, or an agent stack operating in a commercial environment
- Enough time inside frameworks like LangGraph and PyTorch to have opinions on where each one breaks
- Depth in generative AI that goes past prompt engineering, including retrieval systems you have tuned yourself and models you have fine tuned
- Keen eye for detail - email me directly and mention something about the brownie prompt on Harold's LinkedIn profile.
This Role Will Interest You If:
- You've been exploring how AI agents can improve engineering velocity and have found that journey exciting thus far
- You build things because you find the work engaging, not because you were told to
- You do your best work when a team is small, fast, and light on process
If you'd like to apply for this role but are unsure about your suitability, please reach out! With over 20 years of combined experience recruiting for the QLD technology market, VALE Partners love to help you see that you are suitable or assist you in any other way we can.