The Sequel
Two weeks ago I shared that I built an AI product manager so I could stop being one. (Original post)
Here's what happened since.
The team started using it. Not because I told them to. Because it was useful.
Real Conversations, Real Outcomes
One of our sales engineers used it to draft a full UEBA position paper for a prospect in Indonesia. Not a vague outline. A complete technical document covering data flow architecture, behavioral baselining algorithms, risk scoring formulas, MITRE ATT&CK mappings, and detection scenarios. Then he asked the AI to fact-check itself against our actual codebase. It came back and said "we're partially over-committed on 3 of the 6 scenarios." Flagged exactly which detectors exist and which don't. Recommended presenting those three as roadmap items instead of delivered features.
That honesty would have taken me half a day of code review to confirm. The AI did it in under two minutes.
Same day, another sales engineer needed to scope a 65TB log storage project. The AI walked through our hot and cold storage architecture, produced pricing estimates across both tiers, then ran a competitive analysis against Splunk, QRadar, LogRhythm, Logz.io, and SolarWinds. Within minutes it had a full pricing comparison ready, positioned our strengths against each competitor, and drafted a professional response note to the partner. Work that would normally take a day of research and spreadsheet wrangling, done in one conversation.
One of our deployment engineers was planning dashboard implementations for a new deployment. The AI mapped every requirement to our React framework's widget types, identified which data sources each panel would pull from, flagged a missing dashboard that was in scope but absent from the requirements document, and produced a panel-by-panel layout spec with targeted questions for the customer.
Not What I Expected
When I built AMA, I thought I was cloning my product knowledge. What actually happened is something I didn't expect. The AI became a second me for every person on the team. Not a watered-down version. A version that remembers every detail of our codebase, never has a bad day, and is available at any hour.
We're a small team. But right now, each person is operating with the equivalent of a senior architect sitting next to them, one who knows every corner of our product and never needs to context-switch.
The Energy
I'm not going to pretend we have this all figured out. We don't. But the energy in this team right now is something I haven't felt in years. People are excited. They're sharing prompts with each other. They're pushing the AI further every day. And the output is genuinely making us faster in ways I didn't think were possible six months ago.
The tools I built aren't replacing anyone. They're multiplying everyone.
NetGain Systems builds enterprise observability software. Over two decades of domain knowledge, now accessible to every team member through AI.
Top comments (0)