Kit Oliynyk

Proof without the paperwork

A Fannie Mae commercial-lending tool that decides who is in good standing to borrow. The old version was a paper form and a folder of exceptions. The new one took billions in potential risk off the table, and it shipped ahead of schedule because the business and the design were facing the same direction before anyone built a thing.

Client
Fannie Mae. ACheck verifies whether a loan applicant, a person or a business, is in good standing. Fail the check and you are nominated into the ACheck database, which makes you ineligible for a Fannie Mae loan.
Duration
2024. Federal timeline, delivered ahead of schedule.
Role
Design leadership, product strategy and management, business alignment, information architecture, narrative. I led the work.
Team
Four: a researcher, a UX/UI designer, a business architect, and a technical architect.
The ACheck nomination detail: query results, activity feed, approvals

Outcomes

  • 3× savings in development cost, from reusable components and decisions made once, with the business in the room.
  • Delivered ahead of schedule on a federal timeline.
  • Billions in potential risk taken off the table by verifying applicants accurately.
  • FHFA oversight requirements met.
  • A documented design system, components and tokens mirrored in Figma and Angular.
  • A 9.3 average stakeholder rating, which on a federal compliance project is its own kind of win.

Start where it hurts

They were great at listening to exactly what our business needs and asked a lot of questions to make sure we were all on the same page.

Senior Director, Fannie Mae

The whole thing ran on a paper form. Slow, error-prone, and one accident away from losing the data entirely. ACheck 1.0 had digitized it but kept most of the pain: no way to search nominations, a clumsy creation flow, and no automation for approvals or notifications. So that is where we started, with the form that hurt the most, not with the pretty parts.

At Fannie Mae, money does not open up until the architecture is approved, so that is where the design work began too: context diagrams, user flows, and a clear set of roles, with a plain-language proto-persona for each. What hurts, what they want, and what they actually need from the tool, which is usually not the same thing. Doing the architecture first is not the fun part. It is the part that gets the project funded and keeps the whole team pointed at the same problem.

The original ACheck paper form
The original paper form.
ACheck system architecture and context diagram
System architecture and context.
ACheck data flows and dependencies
Flows, roles, and dependencies.

Design with them, not at them

Seeing UI designs while walking through our process has resulted in a more efficient design process.

Senior Associate, Fannie Mae

Everything research surfaced went onto a wishlist. We ran workshops to rank it and pulled stakeholders in early, so the business and the design were facing the same direction before anyone built a thing.

Then we built low-fi prototypes. Not finished screens, just conversation starters. They let us sit down with business stakeholders and design in real time, making decisions with them instead of presenting to them and hoping for the best. We used the same prototypes to check feasibility with engineering early, so the MVP was both something people wanted and something we could actually build. Getting both at once is rarer than it should be.

Ideation and prioritization workshop artifacts
The wishlist, ranked with the business in the room.
Ideation artifacts
Deciding together, out loud, not in a deck.
Low-fidelity prototype
Low-fi as a conversation starter.
Low-fidelity prototype
Search a nomination, the thing 1.0 could not do.
Low-fidelity prototype
The creation flow, rebuilt with the people who use it.

One language, engineers included

Once the concepts held up, we moved to high fidelity, leaning on a growing design system to keep everything consistent. Every component and token was documented in Figma and in Angular, the same names on both sides. We ran onboarding sessions with engineering so we were all using the same words for the same things, nothing lost in translation.

Through build, my team stayed close to the DATS process, checking that what landed in the lower environments actually matched the design and the acceptance criteria. Design does not end at handoff. It ends when the thing in production is the thing you drew.

High-fidelity ACheck interface
High fidelity, built on the documented system.
High-fidelity ACheck nomination detail screen
Nomination detail, with the workflow the old form never had.

What shipped, and what I carried forward

This team is better than any “system development” team with whom I worked on these other major prior projects.

Advisor, Fannie Mae

The app landed ahead of schedule and met the FHFA's oversight requirements, taking billions in potential risk off the table by verifying applicants accurately. Stakeholders were genuinely happy, a 9.3 average rating, and they valued the rapid prototyping, the documentation, and the reusable components, which is where the 3x development savings came from.

Then we carried the lessons straight into the next one. On Property Check we swapped traditional low-fi for AI-powered prototypes, and it works even better: the conversation starts higher up, and the business argues with something that already moves. Same principle, sharper tools, and the spec-and-hope handoff stops being the way work gets lost.