Building PostSign: from contract PDF to trackable obligations in four weeks

Most contract tooling stops at the signature. The PDF gets filed, the obligations inside get forgotten, and the only signal that something is due is when it is already late.

PostSign starts where those tools stop. It takes a signed contract, extracts every obligation, deadline, and condition from it, and turns that into owned, trackable work. I joined as Technical Lead for the AI Start program at Campus Founders, a four-week intensive for early-stage AI startups. This is what those four weeks looked like.

PostSign final pitch at Campus Founders AI Start program, presenting to a full audience of founders, investors, and judges — PostSign at the final pitch, March 31, 2026. Campus Founders, Heilbronn.

The product

A contract lands in PostSign as a PDF. Automated AI extraction pulls out every obligation, deadline, counterparty, and condition as structured data. Before any of that moves forward, a human reviews and confirms the extraction. That review step is not optional: it keeps the system accountable rather than just fast.

PostSign solution architecture slide showing four stages at Campus Founders AI Start finals — The four-stage workflow: secure upload, human-in-the-loop review, active obligation tracking, workspace dashboard.

From there, each obligation becomes a task: an owner, a deadline, a status, and an alert if it goes overdue. The workspace dashboard gives the full picture across all active contracts at once.

PostSign live product demo shown on screen at Campus Founders AI Start program — Live product demo during the final presentation.

PostSign product demo walkthrough.

How we worked

The AI Start format is intentionally intense. Four weeks, weekly check-ins with the Campus Founders team, and a final pitch in front of judges, investors, and startup founders. That structure forces you to be deliberate about what you actually build.

My role as Technical Lead was to make sure we were always building the right thing in the right order. Week one went into a product requirements document and sprint planning. We broke the product into units that could ship independently, so we always had something working even as features were incomplete.

The discipline that mattered most was scope. Every session, something had to be cut. The question was always: does this help us ship a better product, or does it just make us feel productive? Those are not the same thing.

Round 2 results board at Campus Founders AI Start showing PostSign advancing — PostSign advancing through Round 2, and with Marina Iantorno, the AI Start lead at Campus Founders.

PostSign team with Marina Iantorno, AI Start lead at Campus Founders, after advancing to finals — PostSign advancing through Round 2, and with Marina Iantorno, the AI Start lead at Campus Founders.

The outcome

PostSign placed first in the first two rounds and reached the final pitch in front of a room that included judges from the startup and investor community.

Competing in a cohort of AI startups, each with real products and real pitches, was a useful stress test. The problem PostSign solves, contract obligations that go untracked, exists in every industry that signs agreements. That audience understood it immediately.

Campus Founders AI Start Batch 5 full cohort at final pitch day celebration — The full AI Start Batch 5 cohort at Campus Founders.

The team: Bilal Muhammad (co-founder) and me, with Marina Iantorno, the AI Start lead at Campus Founders, providing the program structure that kept us honest about what we were building.