The creator of Clawd: "I ship code I don't read"

Key takeaways

AI tools like Claude and Codex significantly alter the software development workflow, allowing for faster iteration and increased productivity.
The emphasis on traditional code reviews is diminishing as developers move towards using AI prompts to guide coding and collaboration.
Closing the feedback loop between coding and testing is essential for effective AI integration in software projects.
The role of engineers will evolve, focusing more on architecture and system design rather than on writing code.
Companies that adapt to AI technologies will thrive, but this will require a cultural shift towards embracing innovation and flexibility.

Summary

Peter's Journey in Tech

Peter Steinberger shares his tech journey, starting from tinkering with computers in childhood to creating PSPDFKit, a widely used PDF framework. He discusses the challenges of working remotely and the pressures of running a successful tech business, leading to his burnout and three-year hiatus.

The Evolution of Software Development

The discussion moves to the changes in software development practices with the introduction of AI tools. Peter mentions how he has shifted from traditional coding practices to relying on AI agents like Claude and Codex. He emphasizes the need for a feedback loop and the importance of prompts in enhancing productivity.

Agentic Engineering and Workflow

Peter describes his current workflow using AI agents, which he refers to as 'agentic engineering.' He emphasizes how these agents assist him in coding, debugging, and feature development, allowing him to work faster and innovate more creatively.

Shifting Mindset and Company Culture

The conversation touches upon how the traditional engineering judgment shifts when integrating AI into workflows. Peter argues that while quality remains essential, the approach to collaboration and testing evolves, and this requires a cultural shift in how teams operate.

The Future of AI in Software Engineering

Peter shares his optimistic view on the future of AI in software engineering. He believes that as tools become more efficient, companies will need fewer developers who possess high agency and competency. He discusses the implications for new graduates entering the tech field.

Show notes

Brought to You By:

• Statsig — ⁠ The unified platform for flags, analytics, experiments, and more.

• Sonar – The makers of SonarQube, the industry standard for automated code review

• WorkOS – Everything you need to make your app enterprise ready.

—

Peter Steinberger ships more code than I’ve seen a single person do: in January, he was at more than 6,600 commits alone. As he puts it: “From the commits, it might appear like it's a company. But it’s not. This is one dude sitting at home having fun."

How does he do it?

Peter Steinberger is the creator of Clawdbot (as of yesterday: renamed to Moltbot) and founder of PSPDFKit. Moltbot – a work-in-progress AI agent that shows what the future of Siri could be like – is currently the hottest AI project in the tech industry, with more searches on Google than Claude Code or Codex. I sat down with Peter in London to talk about what building software looks like when you go all-in with AI tools like Claude and Codex.

Peter’s background is fascinating. He built and scaled PSPDFKit into a global developer tools business. Then, after a three-year break, he returned to building. This time, LLMs and AI agents sit at the center of his workflow. We discuss what changes when one person can operate like a team and why closing the loop between code, tests, and feedback becomes a prerequisite for working effectively with AI.

We also go into how engineering judgment shifts with AI, how testing and planning evolve when agents are involved, and which skills and habits are needed to work effectively. This is a grounded conversation about real workflows and real tradeoffs, and about designing systems that can test and improve themselves.

—

Timestamps

(00:00) Intro

(01:07) How Peter got into tech

(08:27) PSPDFKit

(19:14) PSPDFKit’s tech stack and culture

(22:33) Enterprise pricing

(29:42) Burnout

(34:54) Peter finding his spark again

(43:02) Peter’s workflow

(49:10) Managing agents

(54:08) Agentic engineering

(59:01) Testing and debugging

(1:03:49) Why devs struggle with LLM coding

(1:07:20) How PSPDFkit would look if built today

(1:11:10) How planning has changed with AI

(1:21:14) Building Clawdbot (now: Moltbot)

(1:34:22) AI’s impact on large companies

(1:38:38) “I don’t care about CI”

(1:40:01) Peter’s process for new features

(1:44:48) Advice for new grads

(1:50:18) Rapid fire round

—

The Pragmatic Engineer deepdives relevant for this episode:

• Inside a five-year-old startup’s rapid AI makeover

• When AI writes almost all code, what happens to software engineering?

• Why it’s so dramatic that “writing code by hand is dead”

• AI Engineering in the real world

• The AI Engineering stack

—

Production and marketing by ⁠⁠⁠⁠⁠⁠⁠⁠https://penname.co/⁠⁠⁠⁠⁠⁠⁠⁠. For inquiries about sponsoring the podcast, email podcast@pragmaticengineer.com.

Get full access to The Pragmatic Engineer at newsletter.pragmaticengineer.com/subscribe