We just launched Paxel!
Paxel is a newly launched tool that analyzes Claude and Cursor coding sessions locally to generate a personalized 'builder profile' across five dimensions. It runs in Docker, keeps code private, and is free to use. Y Combinator is also integrating Paxel tokens into their Startup School application process.
Summary
The transcript introduces Paxel, a tool designed to help developers understand how they build software using AI coding agents. The speaker opens by noting that while everyone is now writing software with AI, the practice is largely uncharted — people are figuring out best practices in isolation without visibility into their own patterns or those of others.
Paxel addresses this by reading a user's Claude and Cursor sessions to surface behavioral insights: whether someone is an architect who plans before building, a late-night shipper, how many agents they run in parallel, and what their go-to prompts are. The tool generates a 'builder profile' across five dimensions — steering, execution, engineering, product instinct, and planning — along with a 'growth edge' that suggests specific next steps based on actual session data.
A key feature highlighted is privacy: Paxel runs locally inside Docker via a single command, meaning code never leaves the user's machine. The profile is delivered to the user's inbox within 15–30 minutes and is free.
The speaker then reveals an integration with Y Combinator's Startup School: applicants can paste their Paxel token into their application to show how they build, not just what they've built. The speaker emphasizes this can only help, not hurt, an application, and encourages those who already applied to go back and add it.
The transcript closes with a broader philosophical point: historically, software development was an exclusive 'priesthood,' but AI has opened the doors. Paxel is framed as a discovery mechanism for identifying talented builders — 'cracked builders no resume would ever surface' — and as a collective experiment to define what great AI-assisted building looks like.
Key Insights
- The speaker argues that despite widespread AI-assisted coding, no one has yet defined what 'building well' with coding agents looks like — everyone is figuring it out independently and in the dark.
- Paxel generates a builder profile across five specific dimensions — steering, execution, engineering, product instinct, and planning — framing it explicitly as 'a mirror, not a grade.'
- The speaker claims Paxel runs entirely locally inside Docker, so the user's code never leaves their machine, positioning privacy as a core product feature rather than an afterthought.
- Y Combinator is accepting Paxel tokens as part of Startup School applications, with the speaker asserting it can only help — not hurt — an applicant's chances, and encouraging those who already applied to retroactively add it.
- The speaker frames Paxel as a talent discovery mechanism, arguing that traditional resumes cannot surface the best AI-era builders and that the tool is an experiment to collectively define what great AI-assisted building looks like.
Topics
Transcript
[0:00] Something strange is happening right now. Every one of us is writing software with AI, but we're doing it alone. No one really knows what it means to build well with coding agents just yet. We're all figuring it out in the dark. So, we built Paxel to turn the lights on. Paxel reads your clawed codecs and cursor sessions and shows you how you actually build. Are you the architect planning first? building scaffolding that compounds, or are you the night owl [0:31] shipping at 1:00 a.m.? Maybe both. It surfaces your go-to prompt, how many agents you run in parallel, and the moves that are uniquely yours. You get a builder profile across five dimensions: steering, execution,…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Y Combinator
AI Can't Learn The Way Humans Do - This Could Fix That
The video explains how world models—neural networks that predict future states of an environment given actions—could solve AI's sample efficiency problem and potentially unlock AGI. By learning to simulate environments like humans do, AI systems could train policies on synthetic data rather than requiring millions of real-world examples, as demonstrated in recent robotics and gaming applications.
New Ways To Design With AI Tools
E Bufar, head of design at Y Combinator, demonstrates how she uses AI tools like Claude and Aqua to design products end-to-end, eliminating traditional design workflows. She walks through three projects—Paxel, Sodazine, and Startup School—showcasing techniques like voice-to-code development, comprehensive markdown documentation, and rapid iteration through disposable design tools.
See How Customers Actually Use Your Product
The speaker explains dot plots, a visualization tool that displays individual user behavior over time by plotting user actions on a 2D grid. Unlike aggregate metrics like DAU/MAU, dot plots reveal usage patterns, feature adoption, and retention issues that are invisible in traditional dashboards.
How A Prototype Built During A Missed Flight Became A New Gusto Product
Eddie Kim, Gusto co-founder and head of technology, discusses Gusto Co-founder, an AI product that automates business processes for small business owners. Built in 10 weeks by a 5-person team during a missed flight layover, the product lets customers automate tasks like payroll and scheduling through SMS and Slack, while proactively suggesting optimizations and uncovering hidden opportunities like tax credits.
India Can Create The Largest AI Companies
Y Combinator leaders and Indian venture investors discuss why India is positioned to create the world's largest AI companies, emphasizing that success now depends on technological edge rather than market knowledge, and that AI has democratized startup building for young founders with high agency and customer obsession.