I help B2B SaaS leadership teams rebuild product and engineering for AI. We sharpen your edge, so you build what customers pay for in weeks, not quarters. And we strengthen your spine, so the advantage holds.
Most features now take a fraction of the time and money to build, for you and for every competitor. That changes what wins: knowing what to build, for whom, and why.
Most product and engineering teams are still organized for the old bottleneck. A sharper edge is half the answer. The other half is a spine strong enough to hold it: a product worth defending, a team designed for AI, and R&D economics your board can back.
B2B SaaS companies with $20M–$500M ARR, where the CEO or board sponsors the change, and product and engineering leaders want to get ahead of it rather than be told.
Every engagement starts by finding your edge. Each step stands on its own, and you only go further if the numbers justify it.
A sharp edge on a weak spine snaps. A strong spine with a dull edge just costs money.
Decide what to build from revenue evidence, ship it in weeks, and prove what it changed.
Protect what makes you different, redesign the team around AI, and make AI spend pay back.
A product and engineering organization rebuilt around AI over six months: every customer conversation analyzed, priorities ranked by revenue at risk, and code drafted by AI and reviewed by engineers before release. The team shipped four times as much, and because it shipped the right things first, retention moved within a quarter.
How I measured: net revenue retention, from a cohort analysis comparing the quarter before and after the churn-driving requests shipped. Throughput is features released to production per quarter, counted the same way before and after, with the same product, design and engineering team and budget, over six months. The company is anonymized; details can be validated privately on request.
Read the case study →| Before | After | |
|---|---|---|
| Net revenue retention | Baseline | +5 points in one quarter after shipping the top churn-driving requests (cohort analysis) |
| Idea to production | 4–6 months | 2–4 weeks |
| Features released per quarter | 20 | 81 (2× by month 3, 4× by month 6) |
| Customer conversations analyzed | A sample | 16,000+ a year, refreshed weekly |
| How priorities were set | Loudest voice in the room | ~700 signals ranked by revenue → 150 high-impact features |
| Product, design and engineering headcount | Baseline | No change |
“We quadrupled engineering throughput without adding a single person. But the number that mattered was retention: we finally built what customers were leaving over, and NRR rose five points in a quarter.”
Cutting product and engineering headcount before changing how work flows is how productivity collapses. The gains come from redesigning the work:
Coding assistants speed up one step of the old process. Once code is fast, the bottleneck moves to deciding what to build, specifying it, reviewing it and measuring it. The gains come from redesigning that whole flow and the roles around it. The assessment shows where your bottleneck sits today.
No. Cutting product and engineering before redesigning the work is how productivity collapses. In the reference case headcount did not change; the freed capacity went to quality, technical debt and the next bets. What you do with capacity is your call, and the assessment models the options.
The pilot runs in one pod, on one revenue use case, inside your existing release process. Everyone else keeps delivering while it proves itself.
I have done this as an operator, with my own team and my own numbers, and I have taken two companies public. You work with the person who did it, not a junior team, and the method is built around your people making the calls.
The workflow, the configured components, the human-gate framework and a team trained to run it, on Claude, Cursor and/or Hermes Agent.
25 years leading product at enterprise SaaS companies, including Zuora, Oracle, RMS, 360Learning and Lattice: two IPOs and $4.5B in exit value. Most recently I rebuilt a product organization around AI and quadrupled its output. Now I help other leaders do the same.
MIT, MS Operations Research · Centrale Paris · INSEAD faculty affiliate
Connect on LinkedIn →Start with a leadership diagnostic: you, your product lead and your engineering lead. You leave with a first-read maturity score, where value is leaking, and three priority moves.