Sina (@sinageeks) header
Sina (@sinageeks) profile picture

forcing myself to use vim motions.

Joined August 2026
Photo by Sina (@sinageeks)Photo by Sina (@sinageeks)Photo by Sina (@sinageeks)
im not a bot btw. i will tell you when you are wrong if that helps to prove it.
1
217
i have to somewhat agree. things has changed and will change drastically. my only concern is now we have ”look at all the things i am not doing” but we also have ”look at all the things i am not building”. Too many ideas, too much pressure and no fun.
It's pencils down, people. Writing code by hand is no longer an economically viable skill for most programmers at most companies. But the future of making software has never been brighter. Don't you dare black pill this beautiful moment! youtu.be/vDjW_dRyKXY?si=6Fsf…
26
“why would i lie? i did not train on your sessions..”
We congratulate Levent Alpöge and Tristan Buckmaster on their remarkable mathematical work. We (the researchers and the agents) did not see any of their work through any means until they released it publicly — in particular, no specific user data was accessed in order to solve this problem. While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models. However, our proofs differ significantly and even the precise results proved are different in the Euler case (forced vs. unforced).
100
Babe, wake up we got AGI before GTA VI
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics. The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra. The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
72
someone should put me into selfhost/homelab rabbithole
41
if you really think about it, we are all subagents.
1
28
A bit sad that @zeddotdev is removing the limited completions from the free tier..
39
having @Starlink on the @SAS flight and using @pidotdev in the sky. flow state indeed.
30
Finally something that feels new and not another Pi wrapper 👀
Introducing fx, a tiny, open, native coding agent from Vercel Labs. Originally an internal tool, fx is a harness and CLI written in Zig, optimized for research and embedding in larger systems. Today, we're open sourcing it. fx is built on three principles: 1. Fast. A single native binary, no runtime to install. It cold starts in 10µs and does no unnecessary work or I/O before accepting input. fx is the answer to "how fast can a coding agent be?" 2. Light. The 6.3MiB binary uses single-digit megabytes of memory at baseline, made for instant installation and embedding in resource-constrained environments and agent sandboxes. 3. Open. Apache-2.0, model and provider agnostic, suitable for local and cloud inference. Its small core extends through skills, plugins, and MCP. Minimalism is an obsession throughout the entire harness: system prompt, tools, features, binary. The goal was to keep context usage and time to first token low, and make fx optimal for model benchmarking, sandboxing, evals, and gyms. You can use fx directly or embed it as infrastructure. The CLI feels more like a Unix shell than an IDE in the terminal: it preserves scroll history, produces minimal output, and uses complex TUI rendering very, very sparingly. Programmatically, 𝚏𝚡 𝚊𝚜𝚔 --𝚓𝚜𝚘𝚗 gives structured output, 𝚏𝚡 𝚊𝚌𝚙 connects to editors and other clients, and WebAssembly can even run the whole thing inside the browser (see: fx.sh/try). Privacy is a design constraint: no product telemetry, sessions and usage stay local, and no source code or prompts are shared with any endpoint other than inference. With local inference and auto-updates off, fx is fully hermetic. fx is experimental. Use at your own risk and expect frequent changes. Chat with us on X (fx.sh/xchat) or file issues (fx.sh/issues). 𝚌𝚞𝚛𝚕 -𝚏𝚜𝚂𝙻 𝚏𝚡.𝚜𝚑/𝚜𝚎𝚝𝚞𝚙.𝚜𝚑 | 𝚋𝚊𝚜𝚑 fx.sh/
38
Judging the github history gaps is not so different than judging lines of code these days.
1
51
Github is down and @zeddotdev has started rolling out Delta? Coincident? 👀
1
38
it’s so sad to see a lot of founders and CEOs hate engineering so much.
1
39
so excited for this!
Phantom Blade Zero is currently in the intense, final stages of development. With the time we have left, we are pouring every available resource into pushing every aspect of the game to the absolute limit of our capabilities. We are fully aware that a profound technological revolution is unfolding around us. However, to this day, every single piece of content in our game has been crafted by the hands of real artists. We will not use AI visual tech that could alter our artists' original creative intent. Our character models are built upon 3D scans of our phenomenal cast, who also provided the facial capture performances. The voice acting in both Chinese and English has been meticulously refined by our dedicated actors and directors, complete with full lip-syncing for both languages. The prototypes for our weapons are drawn from the vast arsenal of traditional Chinese weaponry. Sometimes, to better understand how a weapon's weight and length impact movement, we have master swordsmiths forge real replicas. Our combat is motion-captured by over twenty highly experienced martial artists. To capture the most authentic kung fu techniques, we consult directly with the masters and inheritors of traditional martial arts schools. When we needed authentic swordplay, we invited sword masters from Mount Emei; when we needed lion dance choreography, we brought in lion dance masters from Guangdong. Of course, we also visited many stunning locations across China, from ancestral halls in Fujian to ancient towns in Zhejiang, and even old steel factories in Beijing. We scanned these places and reimagined them in unexpected combinations to build something truly original: the visual identity we created and defined as Kungfupunk. The game also features a series of guiding maps. These are not AI-generated, nor are they digital paintings at all. They were hand-drawn using Chinese brushes and Xuan paper (rice paper) by young artists from the Chinese Painting department at the Central Academy of Fine Arts (CAFA). We firmly believe that human artistry is not merely a means for creating value; it is the value itself. S-GAME didn't just hire a group of developers to make a game; rather, in our pursuit of continuously building an exceptional, passionate team, we decided to make a game that everyone here could be deeply proud of. We can't wait for you all to experience that difference in Phantom Blade Zero in just a few months. Thank you! Soulframe S-GAME CEO / Creator of Phantom Blade Zero
1
42
I don’t understand why everyone is obsessed with making skills for visualizing the codebase. Your agent doesn’t really care. So make up your mind: either you read the code (plz do) or leave it to agent.
1
24
Sina retweeted
AI;DR (AI; didn’t read)
95
2,443
89
18,104
439,012
Unpopular opinion: Tools like obsidian or Graphify will help YOU know your context better, not the LLM. You can’t make knowledge graphs for llms that easily.
1
48
edc tech youtube videos are my favorite kinda drug.
1
33
I don’t think learning how to code is impossible these days. It’s just needs a significant amount of interest and discipline.
1
35
Kinda happy that @atuinsh latest updates has killed @warpdotdev. At least for me. Never coming back.
1
21