Job Title
Founding Engineer, Agent Systems
Salary
Not Disclosed
Company Description
Pre-Series A AI-native risk infrastructure startup backed by leaders from Meta, Palantir, and SpaceXAI
Job Description
You will own the agent platform for an enterprise AI startup already generating seven-figure revenue. Your work involves building the orchestration and reliability infrastructure that turns frontier models into high-stakes production features. You will define the technical standard for agentic systems across financial services, healthcare, and regulated technology sectors.
Location
London, UK
Benefits
* Join an elite founding team of Palantir, Oxford, and Stanford alumni at a company that achieved seven-figure revenue within months of launch.
* Work at the extreme frontier of LLM application; the team recently identified and reported critical bugs in Anthropic's core API during production stress-testing.
* Benefit from top-decile London compensation, meaningful EMI-eligible equity, and significant budgets for specialized AI tooling and API experimentation.
What You Will Do
* Architect and scale agent scaffolding including tool use, context management, sandboxing, and prompt-injection defense mechanisms.
* Develop production-grade evaluation infrastructure for fuzzy, high-stakes outputs using dataset curation and LLM-as-judge methodologies.
* Build robust reliability systems including circuit breakers, fallbacks, and prompt versioning to ensure enterprise-grade stability.
Ideal Candidate
* Professional backend experience in TypeScript with at least 1–2 years specifically shipping production-level LLM features and agent frameworks.
* Deep understanding of systems engineering principles like async queues, idempotency, and regression testing under model swaps.
* Proven ability to own AI quality end-to-end, with the technical confidence to define and enforce shipping standards for complex agentic workflows.
#J-18808-Ljbffr