Building pi in a World of Slop — Mario Zechner
AI Engineer3 HN recommenders · recommended in 3 months, 2026-2026
579K views · 6 months ago · talk · news
Mario Zechner explains why he built pi, a minimal coding agent, out of frustration with existing agent harnesses. He also discusses how AI agents are straining open source projects and argues for slowing down.
HN commenters recommend it as a first-hand argument for building your own lean agent harness instead of relying on bloated, vendor-controlled ones.
1:56:20Let's build GPT: from scratch, in code, spelled out.
Andrej Karpathy
8.2M views • 3 years ago
9:20GitHub Copilot CRUSHES Leetcode Interview Questions! 😳
DevOps Directive
266K views • 5 years ago
2:59:08Vibe coding complex changes in Rust
Jon Gjengset
55K views • 1 year ago
3:15:51Guido van Rossum: Python and the Future of Programming | Lex Fridman Podcast #341
Lex Fridman
1.8M views • 3 years ago
39:46JuliaRobotics: Making robots walk with Julia | Robin Deits
The Julia Programming Language
9K views • 8 years ago
2:29:05Cursor Team: Future of Programming with AI | Lex Fridman Podcast #447
Lex Fridman
1M views • 2 years ago
5:58MarI/O - Machine Learning for Video Games
SethBling
11.5M views • 11 years ago
53:01AI slop attacks on the curl project - Daniel Stenberg
Daniel Stenberg
46K views • 1 year ago
48:32Sparks of AGI: early experiments with GPT-4
Sebastien Bubeck
1.8M views • 3 years ago
1:14:32Building Your Own Dynamic Language
Stanford
21K views • 18 years ago
1:03:04A Cheeky Pint with Anthropic CEO Dario Amodei
Stripe
176K views • 1 year ago
17:54Coding Adventure: Ant and Slime Simulations
Sebastian Lague
2.2M views • 5 years ago
33:13"Pixie - A Lightweight Lisp with 'Magical' Powers" by Timothy Baldridge
Strange Loop Conference
30K views • 11 years ago
2:23:57Sam Altman: OpenAI CEO on GPT-4, ChatGPT, and the Future of AI | Lex Fridman Podcast #367
Lex Fridman
6.8M views • 3 years ago
7:157 minutes of Pharo Smalltalk for Rubyists
Avdi Grimm
37K views • 11 years ago
1:07:22Invited Talk - Guy Steele
ClojureTV
39K views • 9 years ago
1:11Infinite Mario AI - Long Level
Robin Baumgarten
1.3M views • 17 years ago
1:10:00The Mental Game of Python - Raymond Hettinger
SF Python
131K views • 7 years ago
30:10The Inside Story of ChatGPT’s Astonishing Potential | Greg Brockman | TED
TED
1.9M views • 3 years ago
![The Mental Game of Python - Raymond Hettinger [synced version]](https://i.ytimg.com/vi/Uwuv05aZ6ug/hq2.jpg)
1:10:00The Mental Game of Python - Raymond Hettinger [synced version]
Tech Ma
24K views • 6 years ago
3 comments from Hacker News
There is a really good video by the author of pi.dev (which I have used to build some of my own harnesses): ▶ this video The basis of the argument is that the labs are constantly pushing updates to their system prompts that are used in claude code or codex, which are exceptionally bloated and change the sand beneath your feet with every update. By rolling your own harness, as long as you keep up with the latest advances and changes in the ecosystem, you capture a lot of the 'control' that LLM-based development feels like it strips from you. Obvious disclaimer that I use pi.dev when I am aiming for consistency, not absolute quality. Custom harnesses on pi are what I ship, claude code is still my fallback if I need to make sure a PR is the highest absolute quality
Mario Zechner's talk "Building Pi in a World of Slop" is a good listen, on the topic of adults in a "world of slop". ▶ this video
In 18min , Mario Zechner , the creator of pi will echo more or less your exact concerns as to why he developed it ▶ this video Enjoy !