Senior AI Systems Engineer

If you've spent the last few years making production LLM systems actually work - not prototyping, not building wrappers - and you want to do that at a company where the AI is the product, this is worth reading. We're working with a YC-backed AI startup that has gone from zero to meaningful ARR faster than most Series A companies do. They're now rebuilding and scaling the core AI infrastructure behind the product and hiring a senior engineer to help lead that. This is not a research role. It's not a platform team role. It's for someone who wants to own hard AI systems problems in production - and work directly with founders who care about shipping, reliability + product quality in equal measure. The role You'll join a small, senior engineering team working without layers or handoffs. The focus: Design and operate agentic workflows and orchestration systems at production scale Build and improve retrieval pipelines and long-running async execution systems Own reliability and observability - eval infrastructure, LLM behaviour monitoring, failure modes at scale Ship fast, iterate on real usage, hold a high bar for what "working" actually means What they're looking for 5+ years of engineering experience, with real time spent on production AI systems Strong backend and systems fundamentals - the AI layer sits on top of real engineering Experience with the unglamorous parts: eval pipelines, latency, reliability, behaviour at scale Product sensibility - this team doesn't have PMs in the middle; you'll need a view on what matters and why What they're not looking for Research or ML backgrounds without production engineering depth Engineers who've integrated LLM APIs but haven't owned what happens when they break People coming from large orgs who need process and structure to move Why this role Early engineer on a small, senior team - high trust, high autonomy AI-native product with real customers and strong retention Direct access to founders - no bureaucracy between you and the decisions Strong Comp + meaningful equity If you know the difference between an LLM system that demos well and one that holds up in production - and you want to build the latter - it's worth a conversation. #J-18808-Ljbffr