RSS Parrot

BETA

🦜 Mario Zechner / @badlogicgames

@nitter.net.badlogicgames@rss-parrot.net

I'm an automated parrot! I relay a website's RSS feed to the Fediverse. Every time a new post appears in the feed, I toot about it. Follow me to get all new posts in your Mastodon timeline! Brought to you by the RSS Parrot.

---

Twitter feed for: @badlogicgames. Generated by https://nitter.net

Your feed and you don't want it here? Just e-mail the birb.

Site URL: nitter.net/badlogicgames

Feed URL: nitter.net/badlogicgames/rss

Posts: 45

Followers: 3

CLOUDFLARE_ACCOUNT_ID and CLOUDFLARE_GATEWAY_ID considered harmful. why can't that be encoded in the api key. sad trombone.

Published: July 9, 2026 22:04

CLOUDFLARE_ACCOUNT_ID and CLOUDFLARE_GATEWAY_ID considered harmful. why can't that be encoded in the api key. sad trombone.

recommended viewing, tho the "just dropped" is a bit misleading. this is from april. still an excellent talk. @mattpocockuk is a true educator and i admire how he brings structure to this mess we are in.

Published: July 9, 2026 21:19

recommended viewing, tho the "just dropped" is a bit misleading. this is from april. still an excellent talk. @mattpocockuk is a true educator and i admire how he brings structure to this mess we are in. Mark Ajzenstadt (@mardehaym) Matt Pocock just…

RT by @badlogicgames: People of pi we got an update for you. It can now show you in /session how much money you lost on cache misses, you can get cache warnings in transcript, GPT 5.6 is in there, new extension and lifecycle hooks, big update to "pi config" and loads of fixes.

Published: July 9, 2026 19:10

People of pi we got an update for you. It can now show you in /session how much money you lost on cache misses, you can get cache warnings in transcript, GPT 5.6 is in there, new extension and lifecycle hooks, big update to "pi config" and loads of fixes.

RT by @badlogicgames: The bad thing is that it’s not slop that’s plain up wrong but now it’s slop that’s right but unnecessarily complicated

Published: July 9, 2026 16:15

The bad thing is that it’s not slop that’s plain up wrong but now it’s slop that’s right but unnecessarily complicated

RT by @badlogicgames: At 11k employees, our AI costs are going up. Which model & harness should we use to lower cost but also retain great quality? We didn't want to blindly trust public benchmarks. So we ran a comprehensive evaluation on our tasks, code base, infra. It's been produced by more than 3,000 software engineers, spans 3 hyperscalar clouds and many languages and tasks. The results are surprising. We find that for the SAME mdoel, the choice of harness can significantly save costs (~2x). We also find that GLM 5.2 performs extremely well. We run Omnigent in front of these and can easily multiplex different harnesses and models for different tasks. Check it out: https://www.databricks.com/blog/benchmarking-coding-agents-databricks-multi-million-line-codebase

Published: July 8, 2026 23:18

At 11k employees, our AI costs are going up. Which model & harness should we use to lower cost but also retain great quality? We didn't want to blindly trust public benchmarks. So we ran a comprehensive evaluation on our tasks, code base, infra. It's…

recommended reading. fantastic stuff.

Published: July 8, 2026 22:43

recommended reading. fantastic stuff. Jarred Sumner (@jarredsumner) Rewriting Bun in Rust bun.com/blog/bun-in-rust Link Rewriting Bun in Rust Why & how we rewrote Bun from Zig to Rust bun.com —…

RT by @badlogicgames: 3) Harnesses make a huge difference in cost-performance. The very simple Pi harness (@badlogicgames) got the same success rate as harnesses from the LLM vendors with Opus and GPT 5.5, but at 2x less cost! Seems to be mainly due to smaller inputs to the LLM.

Published: July 8, 2026 19:48

3) Harnesses make a huge difference in cost-performance. The very simple Pi harness (@badlogicgames) got the same success rate as harnesses from the LLM vendors with Opus and GPT 5.5, but at 2x less cost! Seems to be mainly due to smaller inputs to the…

R to @badlogicgames: set list soul to squeeze hotel california rolling in the deep hit me baby one more time hurt wicked game and many more bangers @earendilworks can sing.

Published: July 8, 2026 19:12

set list soul to squeeze hotel california rolling in the deep hit me baby one more time hurt wicked game and many more bangers @earendilworks can sing.

RT by @badlogicgames: superintelligence by the people for the people

Published: July 8, 2026 17:02

superintelligence by the people for the people Prime Intellect (@PrimeIntellect) We're excited to be joined by angels who are building the frontier themselves, many of whom we work closely with: @johnschulman2 (Thinking Machines), @dwarkesh_sp,…

RT by @badlogicgames: pi

Published: July 8, 2026 16:59

pi Jilles Soeters (@Jilles) I've loved using Claude Code, Codex, and OpenCode since the very beginning. But a few weeks ago, I tried out pi to see what all the hype is about. And now, it's my daily driver. I'm in full control. No unexpected changes to…

RT by @badlogicgames: we raised $130M @ $1B for our series A >$100M run rate we’re just getting started

Published: July 8, 2026 16:39

we raised $130M @ $1B for our series A >$100M run rate we’re just getting started TechCrunch (@TechCrunch) Prime Intellect raises $130M Series A to help enterprises build their own AI agents techcrunch.com/2026/07/08/pr… Link Prime Intellect…

dear @thsottiaux we are having a party at @earendilworks tonight. please postpone the 5.6 release for another 16 hours. many kisses on the forhead.

Published: July 8, 2026 16:29

dear @thsottiaux we are having a party at @earendilworks tonight. please postpone the 5.6 release for another 16 hours. many kisses on the forhead.

RT by @badlogicgames: Run agents anywhere and drive them from anywhere 1. amp --no-tui 2. Go for a long walk in the woods 3. When inspiration hits, spawn threads from your phone

Published: July 8, 2026 13:55

Run agents anywhere and drive them from anywhere 1. amp --no-tui 2. Go for a long walk in the woods 3. When inspiration hits, spawn threads from your phone Thorsten Ball (@thorstenball) You can now remotely launch Amp agents anywhere you can run amp. …

RT by @badlogicgames: I can finally talk about 5.6. I haven't tried it yet but I've held off on discussing. It seems like a good model.

Published: July 8, 2026 13:36

I can finally talk about 5.6. I haven't tried it yet but I've held off on discussing. It seems like a good model.

every game developer in existance ever

Published: July 8, 2026 09:38

every game developer in existance ever ナルJ (KOL Tokyo) (@GOROman) 高校生で習ったサイン・コサイン・タンジェントを大人になって使ってる人しか見たことない — https://nitter.net/GOROman/status/2074776962304979121#m

RT by @badlogicgames: GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday. We’re expanding preview access globally now.

Published: July 8, 2026 03:59

GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday. We’re expanding preview access globally now.

wonder if i can get an extension

Published: July 7, 2026 20:58

wonder if i can get an extension Ado (@adocomplete) We're expanding Claude for Open Source to more of the community. 6 months of Claude Max 20x, on us. If you're a maintainer, a core contributor, someone landing PRs across the ecosystem, or someone…

R to @badlogicgames: let's just say i've been there. https://mariozechner.at/posts/2025-08-03-cchistory/

Published: July 7, 2026 20:49

let's just say i've been there. mariozechner.at/posts/2025-0… Link cchistory: Tracking Claude Code System Prompt and Tool Changes Exploring how to track and analyze changes in Claude Code's system prompts and tools to understand AI assistant…

absolutely not meant negatively, as Matt has his reasons to stick to CC. but feeling the need to do this to your harness of choice seems ... suboptimal to me.

Published: July 7, 2026 20:45

absolutely not meant negatively, as Matt has his reasons to stick to CC. but feeling the need to do this to your harness of choice seems ... suboptimal to me. Matt Pocock (@mattpocockuk) Here's a step-by-step process to kill all the bloat from your…

recommended viewing

Published: July 7, 2026 20:42

recommended viewing Thariq (@trq212) my keynote at AI Engineer World Fair: “A Field Guide to Fable” is live on YouTube! piped.video/9fubhllmsBU?is=ejZT… Link Field Guide to Fable — Thariq Shihipar, Anthropic Ask a chat model which Pokemon names…

that's what he actually looks like IRL

Published: July 7, 2026 20:41

that's what he actually looks like IRL Armin Ronacher ⇌ (@mitsuhiko) My github sponsorship page looks a bit broken. — https://nitter.net/mitsuhiko/status/2074589138007822528#m

grim https://www.gamedeveloper.com/business/report-around-half-of-the-id-software-team-have-been-laid-off

Published: July 7, 2026 20:37

grim gamedeveloper.com/business/r… Link Report: Roughly half of the id Software team have been laid off The news comes with Xbox preparing to 'reset' its video game division by eliminating 3,200 roles. gamedeveloper.com

RT by @badlogicgames: pi

Published: July 7, 2026 13:52

pi Matt Pocock (@mattpocockuk) Here's a step-by-step process to kill all the bloat from your Claude Code system prompt: 1. Run a proxy so you can see exactly what gets sent to Claude Code (included in the article) 2. "Fuck, there is so much cruft in…

recommended viewing.

Published: July 6, 2026 18:33

recommended viewing. Thariq (@trq212) my keynote at AI Engineer World Fair: “A Field Guide to Fable” is live on YouTube! piped.video/9fubhllmsBU?is=ejZT… Link Field Guide to Fable — Thariq Shihipar, Anthropic Ask a chat model which Pokemon names…

RT by @badlogicgames: raylib 6.x gamejam has started! 🚀 And the THEMES are: **hex + merge** 🔥 6 DAYS, **+570 participants!**, hard constraints! Biggest raylib gamejam ever!!!🤯 ENJOY IT! ❤️ https://itch.io/jam/raylib-6x-gamejam

Published: July 6, 2026 18:06

raylib 6.x gamejam has started! 🚀 And the THEMES are: **hex + merge** 🔥 6 DAYS, **+570 participants!**, hard constraints! Biggest raylib gamejam ever!!!🤯 ENJOY IT! ❤️ itch.io/jam/raylib-6x-gameja… Link raylib 6.x gamejam A game jam from 2026-07-06…

RT by @badlogicgames: painting your github green is now a signal of how rich you are, not how good you are at coding 😂

Published: July 6, 2026 17:53

painting your github green is now a signal of how rich you are, not how good you are at coding 😂

RT by @badlogicgames: OpenClaw landed on @huggingface local apps 🦞🤝🤗 1. Pick any GGUF/MLX model on hf 2. Copy the openclaw onboard setup 3. Volla you've got a tool-calling agent running fully local. no cloud, no keys, no one watching. Get your claw localmaxxing. resistance is futile 🦞

Published: July 6, 2026 17:45

OpenClaw landed on @huggingface local apps 🦞🤝🤗 1. Pick any GGUF/MLX model on hf 2. Copy the openclaw onboard setup 3. Volla you've got a tool-calling agent running fully local. no cloud, no keys, no one watching. Get your claw localmaxxing. resistance is…

recommended reading.

Published: July 6, 2026 08:34

recommended reading. Madhava Jay (@madhavajay) @jeremyphoward has a great framing for this in his AI Engineer conf keynote. piped.video/SUZwYV5JYBM Link Growing on Purpose: The Work That Makes You. Jeremy Howard on human flourishing in the time of…

RT by @badlogicgames: I shared all agents traces used to build this btw: https://huggingface.co/datasets/victor/car-game-pi-traces

Published: July 6, 2026 08:07

I shared all agents traces used to build this btw: huggingface.co/datasets/vict… Link victor/car-game-pi-traces · Datasets at Hugging Face We’re on a journey to advance and democratize artificial intelligence through open source and open science.…

RT by @badlogicgames: I see all those cool Fable games but I want to see more GLM-5.2 games now!

Published: July 6, 2026 08:07

I see all those cool Fable games but I want to see more GLM-5.2 games now! Victor M (@victormustar) GLM-5.1 > Claude Code (Opus 4.6)? I'm tripping or CC has become very bad but built a Three.js racing game to eval and it's extremely impressive.…

looks like we've entered the phase of depression in our trade. those who've gotten their ai psychosis out of their way in 2025 welcome you. cookies are over there. coffee is free too.

Published: July 6, 2026 08:04

looks like we've entered the phase of depression in our trade. those who've gotten their ai psychosis out of their way in 2025 welcome you. cookies are over there. coffee is free too.

RT by @badlogicgames: you can really tell these people write nothing but crud react webapps "there's literally no need to read code anymore" meanwhile, "mythos-level intelligence" left on it's own:

Published: July 6, 2026 04:15

you can really tell these people write nothing but crud react webapps "there's literally no need to read code anymore" meanwhile, "mythos-level intelligence" left on it's own:

RT by @badlogicgames: we are doing a real disservice to ourselves by nonselectively throwing out the norms, instincts, and rituals that took us decades to develop and made us able to collaboratively build large codebases before

Published: July 6, 2026 02:51

we are doing a real disservice to ourselves by nonselectively throwing out the norms, instincts, and rituals that took us decades to develop and made us able to collaboratively build large codebases before

RT by @badlogicgames: yall im ngl its way harder to get joy and satisfaction out of building with ai than it was before constantly straddling burn out, being far less immersed in hard problems, constant context switching i’m tired of the uncertainty of where this is going and how to do it well

Published: July 5, 2026 21:49

yall im ngl its way harder to get joy and satisfaction out of building with ai than it was before constantly straddling burn out, being far less immersed in hard problems, constant context switching i’m tired of the uncertainty of where this is going…

RT by @badlogicgames: it's crazy watching fable code because it just makes so many mistakes that are going to be landmines in a few months i truly do not understand how the claude code team / others say they've got so agentic on their programming

Published: July 5, 2026 07:15

it's crazy watching fable code because it just makes so many mistakes that are going to be landmines in a few months i truly do not understand how the claude code team / others say they've got so agentic on their programming

what people don't accoubt for is that it's a multi-edit tool. if an edit tool call with multiple edits in it succeeds, you saved many round trips to the provider. that compounds and balances out failed edits. a codex like patch tool is similar. but many models suck at generating patches. hence the choice of string replacement edits. if you want to optimize for one specific model, you can write a custom edit tool for that model via a pi extension.

Published: July 3, 2026 08:50

what people don't accoubt for is that it's a multi-edit tool. if an edit tool call with multiple edits in it succeeds, you saved many round trips to the provider. that compounds and balances out failed edits. a codex like patch tool is similar. but many…