Peter Steinberger 🦞
10 Tools · 12 Intel
Peter Steinberger is an Austrian software developer who studied at the Vienna University of Technology and became one of the iOS platform's earliest builders. In 2011 he bootstrapped PSPDFKit, the PDF framework that went on to power document features across nearly a billion devices — reaching $116M in outside funding before he stepped back. After recovering from burnout, he returned to coding in the AI era and in February 2026 joined OpenAI, where he works on agentic engineering.
Is this you? Sign in with X to claim this profile.
Tools
GitHub bot that reviews open issues and pull requests, closing stale or already-implemented ones on strong evidence.
Shared skills, agent rules, and helper scripts synced across machines for Codex and Claude Code.
Intel
Peter Steinberger finds 5.6 Terra high underrated after switching his GitHub review bot to it: about 40% faster with negligible quality loss, better than 5.5 across the board, and far cheaper.
A benchmark pits Codex, Claude, and Grok variants against each other in StarCraft: Brood War across 171 matches, finding none play beyond beginner level, Codex's top variant wins largely through worker-harassment tactics, and Grok spends most turns reasoning without issuing commands.
Peter Steinberger watches Codex drive a browser to upload an image to a GitHub PR comment by hand, because there is no API for it, and runs his agents in VMs so they stop stealing app focus.
Peter Steinberger gives Codex a video-enabled remote KVM to end-to-end test OpenClaw's iMessage integration, since iMessage is unreliable in VMs and read receipts require SIP disabled.
Peter Steinberger's prompt for a full end-to-end QA of OpenClaw: twelve subagents splitting functionality across dev gateways and worktrees, a goal of 200 bugs, root-cause fixes only, and a continuously updated markdown test report.
GitHub sits down with OpenClaw creator Peter Steinberger and core maintainers around the 2.0 release, covering how a project that grew faster than any other on the platform handled scaling pressure and security review as it became infrastructure people depend on.
A practitioner's early-access account of OpenAI's GPT-6 Astra, reporting that the agent chased failures out of the local codebase and into dependencies, producing patches against vitest, tsx and SwiftPM. Useful as a first signal of how far its debugging autonomy reaches.
A Go CLI that saves cameras by name, then captures snapshots and short clips from RTSP/ONVIF devices and local webcams, discovers ONVIF hosts on the network, runs health checks, and fires shell actions on motion. Homebrew install pulls ffmpeg; Windows builds cover network cameras only.
Peter Steinberger explains building OpenClaw, an agent that understands its own source and harness well enough to modify itself, and shares his voice-driven, multi-terminal agentic engineering workflow.
The next version of OpenClaw will use a small decision model to automatically choose whether to steer an in-progress agent turn or queue a new instruction, with support for local and ONNX-based decision models.
Microsoft shipped Autopilot, a persistent proactive agent, built on top of the open-source OpenClaw framework, whose maintainer confirms months of work hardening the codebase for large-scale deployment.
Peter Steinberger previews a coding-agent tooling update that creates git worktrees roughly 80% faster by using native filesystem clone operations (APFS, Btrfs, XFS, ReFS) instead of copying files, also cutting disk usage, with a Rust implementation and a possible port into Codex.