Founding Engineer

We’re looking to bring on a founding engineer who wants to build foundational AI infrastructure, not wrappers.

DAAT is building the context layer for companies. We connect to the tools where work already happens, Slack, Notion, QuickBooks, Google Workspace, and more, then turn fragmented company activity into structured, usable understanding.

The core belief is simple: the missing layer is not more intelligence. It is understanding. Companies already produce enormous amounts of context, but it is scattered, stale, implicit, and hard to reason over. DAAT exists to make that context accessible to people and agents without flattening the reality of the company.

This is a true founding role. You should want ownership, ambiguity, speed, and a lot of responsibility from day one.

• Build core product and infrastructure across ingestion, comprehension, synthesis, memory, and agent interfaces.
• Design systems that preserve source reality, provenance, uncertainty, and tenant boundaries.
• Ship full-stack product surfaces that make complex company context feel simple and usable.
• Talk to users, find bottlenecks, and turn messy operational pain into clean product decisions.
• Work directly on architecture, product direction, and engineering standards.
• Use AI tools heavily, but judge outputs critically and keep the bar high.

• Strong backend and systems instincts.
• Ability to build real product end to end.
• Deep interest in agents, retrieval, context systems, knowledge infrastructure, and applied AI.
• Product-minded taste. You care how the system feels, not just whether it works.
• High agency. You notice problems, propose solutions, and drive things to completion.
• Comfort with ambiguity and founder-speed execution.
• Serious about ideas, systems, and building things that matter.
• Quietly ambitious. High conviction, low ego.

• Experience with TypeScript, React, Supabase, Postgres, Edge Functions, or cloud infrastructure.
• Experience building devtools, internal tools, data systems, search, retrieval, or agent infrastructure.
• Experience with multi-tenant systems, auth, RLS, provenance, permissions, or security-sensitive products.
• Experience using Claude Code, Cursor, Codex, or agentic coding workflows as part of your real development process.
• Startup experience, open-source projects, unusual independent builds, or clear evidence that you can move without permission.

This is not a normal early engineering job.

You will not just be handed tickets. You will help decide what the system is, what the product should become, and how the company thinks about its technical foundation.

The work sits at the intersection of company knowledge, source integration, retrieval, memory, agent tooling, and applied AI. The product only works if the underlying model of reality is honest. That means preserving provenance, uncertainty, source authority, and the difference between raw data, comprehension, synthesis, and memory.

If that sounds exciting instead of annoying, you are probably the kind of person we want to meet.