Case study
How a SaaS team shipped three internal AI tools in two weeks
A VP of Engineering at a SaaS company had engineers spending 10+ hours a week on internal tooling instead of product. We took three of those projects off his plate: an AI research assistant, automated data enrichment and a support ticket router. All three were live within 2 weeks, with code his team owns.
- Trigger
- AI step
- Person decides
- Your systems
- TriggerSupport ticket comes in
- AI stepTicket classified
- Your systemsSent to the right queue, with context
- research assistant, data enrichment and ticket router, all live
- 3 tools, 2 weeks
At a glance
- Client
- A B2B SaaS company, working with the VP of Engineering (anonymized)
- Problem
- Engineers spent 10+ hours a week on internal tooling instead of product
- What we built
- An AI research assistant, automated data enrichment and a support ticket router
- Results
- All three live within 2 weeks
- Time to live
- Within 2 weeks
The client is anonymized. The figures are the real before-and-after numbers from the project, reported as-is and not rounded up.
What was the problem?
Engineers were spending 10+ hours a week on internal tooling instead of product. The projects weren’t big enough to justify pulling a dedicated engineer off the roadmap, but they were too valuable to keep ignoring.
What did we build?
- AI research assistant: gathers and summarizes information so people start from a brief.
- Automated data enrichment: fills and checks records without manual lookups.
- Support ticket router: classifies incoming tickets and sends them to the right queue with context.
All three were built in the company’s own repository and accounts and handed over with documentation.
What changed?
| Measure | Before | After |
|---|---|---|
| Engineering time on internal tooling | 10+ hours a week | Off the team’s plate |
| Internal tools live | Stuck in the backlog | three, within 2 weeks |
- research assistant, data enrichment and ticket router, all live3 tools, 2 weeksSaaS case study: research assistant, data enrichment and ticket router, all live
What should you know before building something similar?
- Batch small internal projects together. One scoping pass across all of them is faster than three separate ones.
- Agree the hand-over standard (tests, docs, owner) at the start, so tools land in a state your team will accept.
- Start the ticket router in suggest-only mode, then let it route on its own once its choices match your team’s.
Could this work for you?
If your team does similar work by hand, probably. Request a free workflow audit and you get a one-page plan: the 3 tasks worth automating first, hours saved, and what a build would cost. See also AI automation for SaaS and Custom AI agents.