Tweet épinglé
Slop to Signal
128 posts
Slop to Signal
@SlopToSignal
Most AI content is slop. I turn it into signal. Better prompts. Sharper posts. Stronger visuals.
California Inscrit le Temmuz 2024
10 Abonnements7 Abonnés
@emollick the artifact isnt the work, the thinking IS the work
nobody built a model for that yet and its 2026
English
A fundamental problem with extending Codex/Cowork/Code to all knowledge work is that they remain very "software-brained" where the end result (the software) is what is important & that code serves as a source of truth.
For a lot of other knowledge work, the process is at least as important as the outcome. This includes researching what is known, an exploration of alternatives, failed efforts, prototype branches, experiments, etc. All of those things are valuable, so you cannot use the PowerPoint at the end the way you can use a codebase, nor is progress on a to-do list sufficient context post compaction. You work in learning loops, refining your perspectives as you go.
In some ways, this makes long-running models like Fable hard to use for deep knowledge work, since they are designed to deliver product to you in the end. You can prompt your way around this problem, but everything about the Codex and Code harnesses want you to be a software developer and you have to fight them. There is a real disconnect between how a manager or analyst thinks about problems and how the agentic software tools approach solving them. Addressing this is critical to breaking out of the coding niche for these tools.
English
@AnatoliKopadze the loop architecture is the actual moat now
smartest model means nothing if ur orchestration is cooked
English
Anthropic engineers just showed how they build a full app from scratch, using a loop of agents
40 minutes from the team behind Claude Code
they used three agents: one to plan, one to build, one to judge, cycling until the app actually works
the winners won't have the smartest model, they'll have the best loop
watch it, then read the full guide on how to actually use loops below
Anatoli Kopadze@AnatoliKopadze
English
@ItakGol the real story is how fast that $20k/month number drops tho
hardware gets cheaper, quantization gets better, give it 18 months
English
This is not another AI slop model. Don’t ignore it.
I spent the last few hours playing with GLM 5.2 after a few friends told me I should pay closer attention.
I was skeptical, as I usually am with open models. Most of the time, I find them way behind frontier labs. Not really comparable.
But today felt different.
I’m not saying it’s perfect, and I didn’t run some comprehensive benchmark suite. But across different domains and tasks, this is the first public open model that felt genuinely close to something like Opus 4.6.
That is a crazy breakthrough.
This feels like a ChatGPT moment for public open models.
The catch: running it properly is still expensive. You probably need something like 8 Nvidia H200 GPUs, which means roughly $400K to buy or around $20K/month to rent.
Not cheap.
But for enterprises currently paying millions per month to Anthropic or OpenAI, this could become a serious disruption.
Open models just got much harder to ignore.
English
@0xfdf the gap between researcher and 'research adjacent' is apparently worth $6M/year and thats the most revealing part of this whole story
English
Temperature check on AI recruiting, here's an n of 1: a friend of mine just left OpenAI ($2M/year total comp) to go to Google ($8M/year total comp, cash/RSUs). They joined OpenAI well past ChatGPT launch and are research adjacent, but mainly software dev, not researcher.
English
@matrosov curious how far this scales before context window starts gaslighting you about what it already found
like finding the same aes impl 3 times and calling it a day
English
Claude Code is a very powerful tool on RE tasks. Looping it with IDA and the decompiler over MCP or idalib delivers very impressive results, especially in combination with dynamic workflows.
/goal find and list all the crypto implementations
[algo => address]
/loop use decompile but verify all the findings with disasm
English
@sentdefender the agency that wasnt even supposed to be using it is the one that got cooked by it
thats not an intel failure thats a whole essay 😭
English
According to The Economist, Sen. Mark Warner (D-VA), vice chair of the Senate Intelligence Committee, stated on June 11 that Gen. Joshua Rudd, director of the National Security Agency (NSA) and U.S. Cyber Command, told him Anthropic's artificial intelligence (AI) model "Mythos" was able to penetrate "almost all of our classified systems" in hours rather than weeks.
The extent of the alleged compromise remains unclear, and no public details have been released regarding how Mythos reportedly gained access or what systems were affected.
The claim is particularly notable given reports that the NSA continued using Anthropic technology in recent months despite President Trump's February directive ordering federal agencies to cease using Anthropic products. It also comes amid recent U.S. government restrictions on Anthropic's ability to export the model, citing national security concerns.
Source: economist.com/briefing/2026/…
English
@mark_k front end devs been out here for 40 years and a model needed like 2 years to catch up
respectfully what took so long 😂
English
Next week will be big for OpenAI Codex 👀
Front end coding finally solved.
Tibo@thsottiaux
We built the Codex App with models that were okayish at front-end. Wait to see what we can do when we finally improve front-end capabilities significantly in our models. That day will be something.
English
@MengTo wild how the 5% still gets its own dedicated tools tho
nobody ever fully divorces cursor lmao
English
Don't sleep on Codex. I've been using it since day one.
- GPT-5.5 xHigh + full access is beast mode
- Computer use + browser use + spawned threads make agent loops incredibly powerful
- Picking up tasks from mobile works surprisingly well
- The new iOS, web, and macOS build plugins are genuinely useful
- I use Codex all day and almost never hit limits
Still using Claude for writing (and waiting for Fable 5), and Cursor when I need raw speed.
But these days, 95% of my work happens in Codex.
Peter Yang@petergyang
I used to be a die-hard Claude Code user. Codex has won me over because: → GPT-5.5 is excellent → Fast mode + generous limits = more reps → Little touches like steering, auto remote control on phone, etc But most of all Codex's browser and computer use capabilities are simply goated. I built so many workflows relying on those two things alone instead of hunting for APIs. I still use Claude Code too. The app seems to be getting better and the design and frontend capability of Opus is still much better than GPT. Whenever Fable comes back that's another reason to go back. Honestly, I hope these two compete forever and other players (Cursor/Grok, Gemini, etc) all stay competitive. This way the builder keeps winning 🙂
English
@4rblaber 'until the tests pass' is carrying this whole pitch
who wrote the tests tho 💀
English
creator of claude code Boris Cherny on why prompt engineering is a dead end:
"we realized early on that asking an llm to write a complex feature in one shot is setting it up for failure. the real magic of agentic coding happens inside the iterative loop."
if you're still manually typing prompts, reading the diffs, and prompting again, you are the bottleneck. top-tier builders don't chat with their tools-they design autonomous feedback loops.
you set the end state. the agent writes, compiles, reads the error logs, and corrects itself recursively until the tests pass.
this 90-minute deep dive is a pure masterclass on how anthropic engineered the ultimate loop-based architecture.
Anatoli Kopadze@AnatoliKopadze
English
@bindureddy opus 4.8 and glm 5.2 and gpt 5.5 in a trenchcoat
weekend drop energy is actually unmatched rn 🔥
English
🚨 AI Agents Can Now Embed Any App And Create Complex Visualizations
Weekend launches are the best!!
- AI agents can embed and use any app
- do detailed system and engineering design
- create 3D models for data-centers and scientific research
We cleverly mix and match LLMs. Opus 4.8 combined with GLM 5.2 and GPT 5..5
Here is a video of the Abacus AI agent designing a 3D data center
English
@BullTheoryio the tool they were using to hack broke into the hackers... thats just poetic
and now nobodys allowed to ask questions about it lol
English
BREAKING: The NSA's own director says Mythos broke into almost all of its classified systems in hours.
Per The Economist, Senator Mark Warner, vice chair of the Senate Intelligence Committee, said General Joshua Rudd, who runs the NSA and the Pentagon's Cyber Command, told him this directly.
This came out on June 11, the same day Amazon reportedly found a separate jailbreak in Anthropic's models. Within hours, Trump ordered Anthropic to cut off foreign access to Mythos and Fable.
Anthropic shut both down completely instead.
Now there are two competing stories for why this actually happened.
One says the shutdown was a response to the NSA's own classified systems getting breached in hours.
The other says Anthropic is privately pushing back, calling the jailbreak minor and the shutdown an overreaction to something other AI models can already be tricked into doing.
The NSA was already using Mythos for its own cyber operations, with Anthropic engineers embedded inside the agency. The same tool the agency was actively relying on is the one its own director says broke into almost everything it owns.
English
@JussiKemppainen fake GI that actually looks good is such a cheat code
unity devs cooking with zero budget hits different
English
day 608: ! built a fake real-time GI system for my game and it cost almost nothing to run! The warm sunset light wrapping under the car is not baked, not a probe volume, not a reflection capture. here is how it works 🧵 #buildinpublic #indiegamedev #unity3d
English
🚀 PayGo is excited to collaborate with @OrixBNB, the AI Agent built on BNB Chain.
Together, PayGo’s request-level payment infrastructure and Orix’s AI capabilities are powering the next generation of Agentic Web3, enabling smarter automation and seamless machine-to-machine transactions.
⚡️ Building a smarter, faster, and more accessible AI + Blockchain future.
English
@EcZachly the exit ramp got removed while people were still on the highway
English
The path of “grind leetcode to get a big tech job and work for 15 years and retire at 35” is dying.
Meta moved half their engineering team to training AI to replace them
Anthropic is a trillion dollar company with only a few thousand employees 1/5
English
@0xMovez 90% of anthropic running agents that improve themselves is lowkey a wild stat to just drop casually
we just out here looping into the future ig
English
Claude Code creator:
"At Anthropic, 90% of our engineers are running agents with self-improving loops.
in 3-6 months, everyone will be running /loops - this is the future of engineering"
in a 1-hour podcast, Boris Cherny reveals the best tips for building Claude Code automations.
Claude + loops + routines + dynamic workflows - that’s the secret.
Watch the talk, then read how to apply and build the same setup in the article below.
Anatoli Kopadze@AnatoliKopadze
English
@RhysSullivan bro it said trust no one, not even the public repo
went full detective mode on itself 💀
English
@_MaxBlade bro the part that gets me is anthropic has whole research teams and salaries and vibes
and someone just uploaded weights 😭
English
@exploraX_ the stirling-pdf one alone just cancelled like 4 subscriptions i forgot i had
my credit card is gonna miss me 😭
English
10 GITHUB REPOS THAT SHOULD BE ILLEGAL TO HAVE.
all free. all open-source. bookmark this for later.
1️⃣ recordly — the free screen studio.
open-source screen studio. auto-zoom, smooth cursor, webcam overlay, styled backgrounds, polished demos without an editor. (AGPL-3.0)
🔗: github.com/webadderallorg…
2️⃣ stirling-pdf — your entire pdf toolkit
self-hosted. merge, split, sign, redact, OCR, convert, compress, 50+ tools, runs locally, nothing leaves your machine. (MIT)
🔗: github.com/Stirling-Tools…
3️⃣ photogimp — turns GIMP into photoshop. photoshop shortcuts, layout, and splash screen patched right on top of GIMP 3+.
🔗:github.com/Diolinux/Photo…
4️⃣ open notebook — self-hosted notebooklm.
drop in pdfs, urls, youtube links — chat with them, summarize, even generate podcasts. bring your own model (18+ providers). (MIT)
🔗: github.com/lfnovo/open-no…
5️⃣odysseus — pewdiepie's self-hosted AI workspace.
chat, agents, deep research, docs, email, memory — local-first, your hardware, your data. (MIT)
🔗: github.com/pewdiepie-arch…
6️⃣freedomain — free domain names for everyone.
register a domain, point it at cloudflare or any DNS, ship your site without paying for the name. (AGPL-3.0)
🔗: github.com/DigitalPlatDev…
7️⃣ hyperframes — write HTML, render video.
heygen's engine that turns html/css + animations into deterministic mp4s. built for AI agents. (Apache-2.0)
🔗:github.com/heygen-com/hyp…
8️⃣ web-to-app — turn any website into an android app, on-device.
configurable webview, apk signing, even node/php/python runtimes — no remote build.
🔗: github.com/shiaho777/web-…
9️⃣ reclip — self-hosted video + audio downloader.
paste a link from youtube, tiktok, x, ig — 1000+ sites — grab it as mp4 or mp3. powered by yt-dlp.
🔗: github.com/averygan/reclip
🔟 excalidraw — the infinite whiteboard that replaces miro, figjam and lucidchart. hand-drawn diagrams, wireframes, real-time collab, end-to-end encrypted. 120k+ stars for a reason. (MIT)
🔗: github.com/excalidraw/exc…
most people pay monthly for tools that already exist for free. you don't have to.
m0h@exploraX_
10 free github repos that can replace major SaaS with subscriptions. all free. open-sourced. some are MIT licensed. — 1️⃣ openscreen — replaces screen studio ($29/mo) - a clean macOS/windows/linux screen recorder for polished demos. - blur, cursor highlighting, annotations, export to mp4 or gif at any aspect ratio. - doesn't try to clone every feature, just nails the basics for quick walkthroughs you'd post on X. — 2️⃣ voicebox — replaces elevenlabs ($22/mo) + wisprflow ($15/mo) - local-first AI voice studio. - clone voices from 3 seconds of audio, generate speech across 7 TTS engines in 23 languages, - dictate into any text field with a global hotkey. - nothing leaves your machine. - runs on apple silicon, cuda, rocm. — 3️⃣ openshorts — replaces opus clip ($19/mo) + submagic ($16/mo) - free AI video platform. - clip generator turns long youtube videos into 9:16 shorts with auto-subtitles and face tracking (runs on free gemini + elevenlabs tiers). - also includes AI UGC video generation with actors — that part is pay-per-use via fal. ai (~$0.65-2 per video). docker self-host. — 4️⃣ freellmapi — replaces chatgpt pro + claude pro ($20/mo each) - stacks 14 free AI provider tiers (google, groq, cerebras, openrouter, github models + 9 more) behind one openai-compatible endpoint. ~800M tokens/month. - smart router with failover, sticky sessions, encrypted key storage. ships with a dashboard. — 5️⃣ playwright-mcp — replaces browserbase ($39/mo) + browser use ($25/mo) - microsoft's official MCP server that gives any AI agent full browser control. - uses accessibility trees, not screenshots — deterministic and token-efficient. - works with claude code, cursor, windsurf, codex out of the box. — 6️⃣ vibe-trading — replaces tradingview premium ($60/mo) - natural-language finance research agent. - 7 backtest engines across stocks, crypto, futures, forex. - 75 specialist skills (factor analysis, options strategy, ML strategy). - 29 multi-agent swarm presets. - 21 of 22 MCP tools work with zero API keys. — 7️⃣ CalCom — replaces calendly ($12/mo) + savvycal ($12/mo) - the open-source scheduling infrastructure. - one-on-ones, group events, round-robin, team booking, - payment collection (stripe), routing forms, workflows. - integrates with google/outlook/apple calendar, zoom, meet, teams. - self-host in 10 minutes with docker. 40k stars. — 8️⃣ whisper — replaces otter ($17/mo) - openAI's open-source speech-to-text model. - transcribe audio in 99 languages, translate to english, generate timestamps. - runs locally on cpu or gpu. - the actual model behind most "AI transcription" SaaS tools you're paying for. — 9️⃣ postiz — replaces buffer ($15/mo) - AI-powered social media scheduler. - cross-post to X, linkedin, instagram, tiktok, threads, bluesky, mastodon, youtube, pinterest. - AI captions and hashtags. - analytics dashboard. team workspaces. 31k stars and rising. — 🔟 vaultwarden — replaces 1password ($8/mo) - unofficial bitwarden-compatible server written in rust. - works with every official bitwarden client (mobile, desktop, browser). - unlimited users, unlimited vaults, full enterprise feature set. - runs on a $5 VPS or your home server. — disclaimer: open-source ≠ 1:1 replacement. you'll trade polish for ownership, hand-holding for control, and a credit card for a github version. for builders, prototypers, and indie hackers — that's the whole point. for everyone else, the paid tools still have their place. bookmark this. share with one friend bleeding subscription fees. ~m0h
English
@ajambrosino bro said the users are wrong actually
the Taste™ trademark is doing heavy lifting here 💀
English
in our internal tool that predated the Codex app, I tried to growth hack by wrapping the upsells in HDR
everyone hated it, but that’s because they have no Taste™
English
@gokulr wild that the people building these things are also routing around their own model's weaknesses
like even insiders dont trust it end to end lmao
English
After using Claude Code (Opus 4.7/4.8) and Codex (GPT-5.5) incessantly for the past several weeks, my verdict is that (like every human) their greatest strength is also their greatest weakness.
tl;dr Use Claude Code + Opus 4.7/8 for brainstorming and planning. Use Codex + GPT 5.5 for execution and building. Use both to adversarially review each other's plans / design docs.
CC is really creative and a great brainstorming partner. However, this creativity makes it hallucinate when executing.
Codex is an incredible, focused, fast executor and builder. However, this makes it poor at generating new, creative options.
(I have friends at both OpenAI and Anthropic who agree with the above and use the "other" lab's product for precisely the use cases that their product is not good at).
English