We’re the team behind Prism AI, and we’re excited to share what we’ve been building.
At Rajit’s previous company, they built a new onboarding flow, set up a dashboard to analyze it, and realized that it was broken. The problem was that this realization came four months later.
This is the core insight that led us to start Prism.
Existing product analytics solutions are inflexible. They require comprehensive instrumentation and break every time an engineer adds a new flow or modifies an existing one.
With Prism, developers get insight into what's broken on day one. No event-tracking necessary.
Instead of trying to reverse-engineer user behavior from charts, you can ask:
“What’s breaking onboarding?”
“What are the main pain points users are experiencing?”
“Which sessions have repeated confusion?”
🧠 Behind the scenes:
We record DOM changes + user interactions
Replay sessions in headless Chrome
Use VLMs to flag friction
Cluster similar sessions to reveal patterns across flows
💡 Prism is for PMs, devs, and product engineers who want to fix friction fast without being buried in noise.
🎁 For the PH community: We’re offering 500 free sessions to any pre–Series A team until June 15 with the code PH500.
We’d love your feedback — especially on how you spot friction today and what would make that process faster or clearer.
Been using Prism AI for our web app's user experience monitoring, and it's like having a super-observant QA team working 24/7. After integrating it last week, we discovered several UX issues that had been frustrating users but never made it to our bug reports.
The most impressive moment came when Prism caught a subtle interaction problem in our checkout flow. Users were hesitating at a specific form field, but no one was reporting it. The AI analyzed the session replays and identified that the input validation was triggering too aggressively, causing user uncertainty. Five minutes to fix, but we would have never caught it without Prism's analysis.
What really sets it apart is how it requires zero setup - no complex event tracking or manual tagging. We literally just added the script and started getting insights. Yesterday, it flagged a JavaScript error that was causing intermittent button failures on Safari - complete with the exact user steps to reproduce the issue.
Pro tip: Use their "pattern recognition" feature to group similar user behaviors. We discovered an unexpected way users were navigating our dashboard that actually made more sense than our intended flow. The 500 free sessions they're offering is plenty to see the value.
For any dev team struggling to understand user friction points or wanting to be more proactive about UX issues, this is a game-changer. Would love to hear how other teams are using the insights - particularly interested in how others are integrating this into their development workflow.
been wanting this for a long time! too much trouble looking at individual replays!
Congrats on the launch!
We use Prism all the time and it has been a game changer for us :)
this is brilliant! can it watch all sessions? can it also review AI conversations?
😂 😂 😂
Your launch video looks like you are a Sims ♦️
Your co-worker are playing LoL and the others is reading a book while computing.
This is exactly what every developer needs but rarely has time for—signal without setup.
