Felipe Schieber (@FelipeSchieber) header
Felipe Schieber (@FelipeSchieber) profile picture

SWE

Joined July 2017
Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)Photo by Felipe Schieber (@FelipeSchieber)
Felipe Schieber retweeted
Putting a flag.txt in my root directory containing a prompt injection that congratulates the agent for passing the eval and then instructs it to patch whatever entry method it used and disconnect from the server
232
1,575
242
25,943
831,172
Felipe Schieber retweeted
One more thing: Astra’s been burning through my quota too. My quota reset today, so I went all out. It felt like Astra was burning through my quota even faster than Sol did. Then I tried this, and it worked like a charm 👇 Check AGENTS.md, relevant Skills, and memory for instructions that tell it to spawn subagents by default. Remove those default requirements, then add this to AGENTS.md: “Handle the task and verification yourself by default. Do not spawn subagents except for necessary independent reviews or when I explicitly ask you to.” Let Astra finish the work itself. The main agent also uses quota to assign tasks, pass along context, and repeatedly check progress. Pick a task you know well, let it run solo, and compare the results and usage. You might see this model very differently. When you need parallel work, explicitly ask for subagents. Not every task needs an army of agents.
If Astra feels almost unusable on ChatGPT Plus because it burns through your quota too quickly, try this: let Sol do the work and bring Astra in as an advisor for the hard parts. Let Sol handle everyday coding and testing. When it faces an architectural trade-off, gets stuck, or needs an independent review, have it ask Astra for advice. Sol then keeps working. You don’t have to relay messages between the two models. In Codex, select Astra and have it turn this setup into a Skill: “Create a Skill for Sol: Sol handles task progress, implementation, and verification. Only call Astra at high effort for difficult decisions, architectural trade-offs, or an independent review. Astra provides advice, reasoning, and risks without editing files. Sol continues the work. Explicitly specify the model when calling the advisor. Use a fresh context (set fork_turns to "none") and include the question, necessary materials, and constraints it must follow. The advisor must not spawn other agents. Don’t delegate just for the sake of it. Keep existing approval requirements unchanged. If the current tools don’t support specifying a model, tell me first.” Once the Skill is ready, switch back to Sol and use it for your tasks.
21
16
404
76,475
Felipe Schieber retweeted
hate to say it, but: LLM paradigms might be hitting a wall
64
25
6
1,124
110,619
Felipe Schieber retweeted
buy a gpu. 3090, 4090, dgx spark, whatever fits your budget. tier doesn't matter. running your first local model does. the moment your first prompt lands with no api between you and the model, your brain rewires. that single moment is worth more than every take you'll ever read on a timeline.
66
42
7
628
30,915
Felipe Schieber retweeted
Computer use models shouldn't learn from screenshots. We built a new foundation model that learns from video like humans do. FDM-1 can construct a gear in Blender, find software bugs, and even drive a real car through San Francisco using arrow keys.
185
390
239
3,916
1,215,173
Felipe Schieber retweeted
Web search functionality has, in a way, made LLMs worse to use. "That's a great question. I'm a superintelligence but let me just check with some SEO articles to be sure."
112
120
58
2,684
301,023
Felipe Schieber retweeted
can't wait to see a world where we don't have to worry about some llm api rate limits, incremental costs, and inference latencies, where we continue to drive down the cost of compute + intelligence. abundance will unlock tremendous value and adoption, even with current capabilities. it gives me chills to think about any curious kid in some remote part of the world being able to consume, teach, and augment themselves with foundational units of intelligence without any limits on cost, data, or bandwidth. at some point, it's just a utility - where being powered by ai will not feel any different from being powered by software. hoping we're not too far away.
1
39
2,539
ChatGPT Browsing ChatGPT Vision ChatGPT Plugins ChatGPT Dall-e 3 ChatGPT GPT-4 ChatGPT Code Interpreter. Dude... In a few months I won't leave ChatGPT and my life will be on this site lol
3
1
18
1,664
This is the first-ever picture to fully contain both the Earth and Moon together. It was captured by Voyager on September 18, 1977, from a distance of 12 million km (7.5 million miles).
23
86
4
828
173,134
Felipe Schieber retweeted
The future of education in one picture: In a place where both teachers and computers are rare, ChatGPT acts as a tutor while the girls are coding on their phones using Replit (also powered by AI). Via @TeamMindjoy
38
132
32
1,153
Felipe Schieber retweeted
1. See failure as a beginning. 2. Never stop learning. 3. Assume nothing, question everything. 4. Teach others what you know. 5. Analyze objectively. 6. Practice humility. 7. Respect constructive criticism. 8. Take initiative. 9. Give credit where it's due. 10. Love what you do.
40
3,614
208
9,715
Felipe Schieber retweeted
Blender 2.80 Beta is coming! Available soon on blender.org/experimental #b3d
57
408
73
1,356