Senior Product Manager · Global B2B SaaS, 22 Countries · Applied AI
AI product judgment, in both directions.
7.5 years building enterprise B2B SaaS for CPG — and taking it past launch to scale: a 22-country platform consolidation covering ~60% of a client's global revenue, a 300K-buyer commerce portal, an agentic AI product now in production. I don't just ship 0→1 — I get products adopted at global scale, and I know when to kill an AI build that isn't worth the adoption cost. Since January 2026, I've been building FieldIQ, a live AI field-selling platform, solo, end to end.
Price-to-Pocket Bridge
How I think about price — the RGM framework, illustrated (index values, not a specific deal)
The record, in numbers
AI Product of the Year
First enterprise deployment plus two follow-on deals.
4 agents
Shipped solo — discovery, PRDs, UX, prompt design, eval, integration.
56 → 1
Legacy tools consolidated across 22 countries. Reps went 4 → 7 calls/day.
300K+
Business buyers, live in under 9 months. 15% CSAT lift, ~18% repeat-order uplift.
~30% / ~20%
Lift in SKU-mix and promo-mix target achievement via computer vision.
1 ML pilot killed
Salesforce Einstein route planner replaced by a simpler system that outperformed it.
Case files
FieldIQ — AI field-selling platform
Built solo, end to end: four production agents for CPG field reps, with a model-agnostic gateway, strict JSON schema contracts, coded guardrails, and a hand-adjudicated eval harness with a ship gate.
Read the case file →
RGM·AI — 2024 Stevie Award winner
Agentic pricing and revenue-growth-management product. Removed three buyer objections with one architecture choice — compose on the customer's own LLM via RAG over their own data.
Read the case file →
Sales Execution 56→1
0→1 mobile platform replacing 56 legacy tools across 22 countries and ~60% of the client's global revenue — plus the adoption crisis that followed, computer vision that worked, and an ML pilot that got killed.
Read the case file →
B2B Commerce & Loyalty Portal
Self-service commerce for 300,000+ business buyers, live in under nine months — including a loyalty design that failed first, was paused mid-build, and was rebuilt around observable behaviour.
Read the case file →
Why product management
The hardest parts of the job were never really technical. They were judgment calls — what to build, what to kill, whose version of the problem to trust. Getting promoted to Product Owner meant I finally got to make that call myself — and that's when I figured out what I actually enjoyed about this work.