@ Zalos
Zalos — Founding Full-Stack Engineer (London, UK; in-office; full-time). Zalos is the AI "action layer" over enterprise systems of record (SAP, Salesforce, ServiceNow): users describe an outcome in plain language and Zalos figures out which systems to touch, in what order, with which data, and executes it with approvals and a clean audit trail — including computer-use agents for the 30-40% of workflows that live in screens/VDI/legacy thick clients with no reliable API. Paying enterprise customers, strong seed funding, small fast team.
You'll design and own the core agent architecture: natural-language intent -> decomposition into steps -> calling the right APIs and UI agents -> graceful failure handling -> closing the loop with the user. You'll build reliable, production-grade LLM pipelines (context management, tool use, structured outputs, multi-step reasoning over real enterprise data), and set the technical standards for how Zalos builds AI. Work directly with the founders and customers.
Must haves: shipped AI-powered features that real users depend on in production (you know what breaks and how to prevent it); deep hands-on experience with LLM APIs, agentic patterns, and tool/function calling; reliability, latency and cost as first-class concerns; builds tests/evals before shipping; high ownership (ambiguous problem -> clean working solution with minimal hand-holding); pragmatic in a startup.
Client preferences: 3-5 years of experience, strong CS foundation, full-stack SDE leaning backend, worked at startups a strong plus. Modern AI/startup pedigree preferred (not legacy/dated-enterprise). London-based, UK work authorization (visa sponsorship considered case-by-case).
Nice to haves: computer-use / browser automation (Playwright, browser agents, UI parsing); fine-tuning or RLHF on domain tasks; prior founding-engineer or early-stage startup experience.