Skip to main content

Case study · Zero-to-one applied AI

Rally

Release candidate

Helping new hires practice how experts think, not just read what they know.

What to look for: the product hypothesis, the system architecture, why scenario practice was chosen and an actual demonstration.

At a glance

Built by
Sunny Patneedi
Rally product film, end card
Stage
Release candidate
Rally README · release status (via How I build)
For
New hires and the teams who onboard them
Product registry
Checked with
Synthetic company demos
Rally README · release status (via How I build)

01Product hypothesis

Onboarding is reading. Expertise is judgment.

  • Onboarding gives new hires information to read. It rarely lets them practice how experts think.

    Source Homepage evidence card for Rally

  • What if onboarding helped people practice how experts think, rather than simply consume information?

    Source Homepage Rally card

  • Write the product contract first.

    A filled brief names the learner, the buyer hypothesis and the first useful result: one realistic company decision, explained with its source. It also names what Rally must never do, such as scoring, ranking or hiring decisions.

    Source Product and design contract (via How I build)

02Why scenario practice

Practice one real decision, with its source.

  • The first useful result is one realistic company decision, explained with its source.

    Source Product and design contract (via How I build)

  • What Rally must never do

    The contract names what Rally must never do, such as scoring, ranking or hiring decisions. AI coaching and human feedback stay separate.

    Source Product and design contract (via How I build); product registry

03System architecture

AI drafts. People publish.

  • AI drafts. People publish.

    The workspace turns reviewed company sources into role-specific lessons and practice. AI drafts carry their sources; a person approves before any learner sees them, and Drive imports re-check each reader’s own permissions.

    Source Rally README · learning workspace (via How I build)

  • The stack

    Interface: React, Vite, Tailwind. Server: Express 5, PostgreSQL. Identity: Clerk. AI: OpenAI via the AI SDK, Provider storage off. Sources: Google Drive, per-reader permission checks. Checks: Vitest, Playwright, Disposable PostgreSQL.

    Source Rally README (via How I build)

  • Let review change the design.

    Independent review found AI coaching text that was too open-ended. Now the model picks from a small set of allowed suggestions and the server supplies the words. Model-written conclusions and scores are rejected.

    Source Verification record · review corrections (via How I build)

04Demonstration

The decision lab, on film.

  • A synthetic demo: a customer asks for a feature by Friday. The learner picks A, B or C, checks a sample source (a fictional customer-commitments policy) and gets one clearer next step. The demo content is authored, not AI output.

    Source Rally product film, on-screen text

05Evidence

What has been checked, and what hasn’t.

  • Test the real path, not the mock.

    51 browser tests drive real browser → Express → PostgreSQL journeys, including revocation, cancellation and retry, on top of 65 foundation tests and 106 application tests.

    Source Verification record, Nov 7, 2026 (via How I build)

  • Name what isn’t proven.

    Rally is a release candidate, not a verified launch. Live provider and hosted acceptance are still open, and the record says plainly that local fixtures don’t prove production works.

    Source Rally README · release status (via How I build)

  • A release candidate shown with synthetic company demos. No public page has been checked, it is not a verified production launch, and there is no evidence yet that it shortens ramp-up time.

    Source Product registry

The film.

Read the transcript

An overhead view of scattered office paper and folders squaring up into one neat stack on a navy desk. Three cards marked A, B and C appear beside a short scenario, a sample source sheet slides over the middle card, and an orange note names the outcome. It ends on a paper end card with the Rally wordmark.

On-screen text: Early product preview. Get everyone on the same page. Make your call. Decision lab · synthetic demo. A customer asks for a feature by Friday. What do you promise? A, B, C. Check the sample source: Customer commitments · v3. Fictional policy. Authored demo content, not AI output. One decision. A clearer next step. Rally. Get everyone on the same page. Early product preview. Built by Sunny Patneedi. Join the waitlist.

What next?

Let’s build something important.

If you’re building something where the technology is promising but the right product is still being discovered, I’d love to talk.

Each statement names its source. Nothing here is a production benchmark unless its source says so.