Staff AI Engineer

uiAgent is building AI agents that are transforming how CPAs and accounting teams work. We're pre-Series A with significant revenue, strong traction, and the kind of product-market fit that makes investors pay attention. The accounting industry is massive, underserved by technology, and ripe for disruption — and we're already doing it.

Our team is small, sharp, and close-knit. We built this product over 18 months by being relentlessly resourceful, and we're looking for someone who operates the same way.

The Role

We need a Staff AI Engineer who can own the intelligence layer of our platform end-to-end — from the agent architectures powering our product to the infrastructure keeping them reliable at scale. At our stage, there's no "that's not my job." You'll move fluidly between model integration, backend systems, and infrastructure because that's what building a real AI product demands.

* AI & ML — You'll design and build the agentic systems at the core of our platform. This means architecting tool-use frameworks, building RAG pipelines, designing multi-agent orchestration, prompt engineering, fine-tuning, and integrating foundation models into production workflows. You'll push the boundaries of what our agents can do autonomously while keeping them accurate and reliable for high-stakes accounting and audit work.

* Backend & APIs — You'll architect and build the services that support our AI systems — microservices in Node.js and Python, scalable APIs, and databases handling complex audit and financial data. You care about clean abstractions, thoughtful schema design, and systems that don't fall over at scale.

* Infrastructure & DevOps — You'll own the infrastructure our agents run on: CI/CD pipelines, containerization, cloud infrastructure on AWS, and observability. When a CPA firm runs 500 audits through our system simultaneously, you're the reason it doesn't break.

This isn't a role where you'll write tickets and hand them off. You'll identify problems, design solutions, build them, ship them, and keep them running.

What We're Looking For

* Production AI experience. You've shipped AI-powered features to real users, not just built demos. You understand the gap between a working notebook and a reliable production system. You've dealt with hallucination mitigation, evaluation frameworks, latency optimization, and the messy realities of LLMs in production.

* Deep understanding of LLMs and agentic systems. You have hands-on experience with tool use, retrieval-augmented generation, multi-agent coordination, and structured output. You follow the space closely and have strong opinions about what works and what's hype.

* Strong engineering fundamentals. You have opinions about code quality, system design, and architecture. You hate spaghetti code because you've seen what it does to velocity at scale. You're comfortable across the stack — backend, infra, and enough frontend to be dangerous.

* Startup tempo. You thrive in environments where speed matters and context-switching is the norm. You're self-directed and don't need a project manager to tell you what's important.

* In-person collaboration. We work together in our Manhattan office. Our best ideas come from whiteboarding sessions, fast feedback loops, and the kind of spontaneous problem-solving that doesn't happen over Zoom.

What We Offer

* Substantial equity — meaningful ownership in a company with real traction and clear upside

* Competitive salary + cash bonus

* Flat structure — you'll work directly with the founders and have real influence on product and technical direction from day one

* Top-tier equipment — whatever you need to do your best work

* Healthcare, dental, vision, and commuter benefits

* Daily team dinners and a culture that actually enjoys spending time together

* A great office in Manhattan with a skyline view

Who We Are

We're a team that takes the work seriously but doesn't take ourselves too seriously. We're direct with each other, we solve hard problems together, and we genuinely like hanging out. If you're looking for a place where your work has outsized impact, where you'll learn fast, and where you're building something people actually use — we'd love to talk.