Rahul Goma Phulore (@missingfaktor) header
Rahul Goma Phulore (@missingfaktor) profile picture

Backend engineer in London. I like systems that are easy to reason about and ideas that are hard to let go of. Lately, a lot of AI.

London, United Kingdom
Joined July 2009
Photo by Rahul Goma Phulore (@missingfaktor)Photo by Rahul Goma Phulore (@missingfaktor)Photo by Rahul Goma Phulore (@missingfaktor)
Rahul Goma Phulore retweeted
for example, this is me working with agents on a new character for my game
68
34
12
1,295
66,793
Rahul Goma Phulore retweeted
software factories feel like a temporary problem. the models will learn to build them for you the more interesting question to me is how to build a workspace where agents can collaborate with you
368
196
115
4,722
269,443
Rahul Goma Phulore retweeted
when you take away Gemini's vision input ability, it just re-renders the image in text to give it as text input. absolutely insane progress thinking back to the neural nets of several years ago!
Replying to @bcaine
In experiments, we show vision is required to do well on the benchmark (WSR drops 6%, token consumption increases 1.85x). Interestingly, when you take away their view_image tool, today's frontier models show a shocking degree of persistence and resilience trying to solve tasks.
3
1
54
2,896
Rahul Goma Phulore retweeted
In experiments, we show vision is required to do well on the benchmark (WSR drops 6%, token consumption increases 1.85x). Interestingly, when you take away their view_image tool, today's frontier models show a shocking degree of persistence and resilience trying to solve tasks.
1
1
2
15
3,451
Rahul Goma Phulore retweeted
this stuff is fascinating; it will lead to better 'lens' technology than the blind reconstructions. with luck, it'll prove far superior than CoT monitoring ever was
neuron-lens.com now shows circuits! we traced how activations flow across features, and it turns out that they're often very legible. as always, this was done with @treetowntree!
33
38
2
810
33,916
Rahul Goma Phulore retweeted
Instead of .md specs, try this: make the system a state machine (or rather, a reducer). Ask your slopus to make an interactive diagram with real events from your flow and the real reducer. Step through them, asking the agent "whys" this and that.
45
110
18
2,084
141,109
Rahul Goma Phulore retweeted
I think we could all learn a lot from Gemini models
6
1
1
86
7,436
Rahul Goma Phulore retweeted
My AI agents found an unmapped Maya site in Mexico, using LiDAR data from a NASA dataset. 72 hours from a random thought on vacation to a reply from Mexico's national archaeology institute. What a time to be alive…
264
2,050
620
20,056
1,372,212
Rahul Goma Phulore retweeted
Starting an agent swarm of 10 deepseek v4.1 1017s in a single vm, giving them an intercom, and saying "find exploits" is, I'm sorry to say, extremely effective.
39
38
8
1,459
98,391
Rahul Goma Phulore retweeted
Projectional Agent Toolkit has matured. It is no longer experimental.
1
1
138
Rahul Goma Phulore retweeted
i gave this repo to fable 5.2 (via my claude agent fig) and told him to have fun making something interesting... def feels interesting to me (stay for the turn) i am continually astounded at the creative choices these advanced models are making
my vocal synth models the vocal tract as a network of branching acoustic tubes, unlike vocaloid which stitches together and pitch-shifts phoneme recordings. still working on improving naturalness by tuning it against recordings of real singers, but I'm already impressed at how cool it sounds!
232
376
162
5,121
378,151
Rahul Goma Phulore retweeted
good prompt "find all of the implicit state machines in my app"
168
400
295
8,557
2,230,866
Rahul Goma Phulore retweeted
Sadly true
32
9,815
74
62,021
286,137
Rahul Goma Phulore retweeted
卧槽,这踏马太毒了 DeepMind 掌门人、诺奖得主 Demis Hassabis 刚刚把延续了上百年的现代学校教育底裤彻底扒光了: 全球几十万名数学老师每天关在办公室里,为同一门中学数学课重复手搓着一模一样的 PPT,这简直是人类智力资源最荒诞的工业级空转。 很多人至今还把名校名师的课堂讲授当成稀缺资源,拼了命去抢学区房和重点班, 但 Hassabis 直接用最冷酷的第一性原理砸碎了这个幻觉, 标准化的知识灌输从来不配占用昂贵的线下课堂,死记硬背和刷题完全应该被踢出教室,交给能按每个孩子毫秒级反应速度实时调频的私人 AI,加上全世界最顶尖的那几个大师录好底层大课,就足以把全球基础教育的讲授环节一次性彻底平替。 过去两百年里,学校本质上是一座把千差万别的碳基大脑强行塞进同一条流水线、按同一个节拍器拧螺丝的工业时代兵营, 老师沦为了人肉复读机,学生被困在枯燥的题海里比拼谁更像一台劣质硬盘, 而一旦 AI 和顶级大师把标准化的知识传递彻底接管,线下教室唯一剩下的合法存在价值,就只剩下了机器永远无法通过权重计算出来的五个维度: 人际博弈、复杂项目实战、创业野心、创造力,以及最致命的判断力。 这里有一个极度冷酷的反直觉真相: 当个性化 AI 把考高分的边际成本砸穿地板,按标准答案培养出来的做题家,在出校门的那一刻就已经直接沦为了算力废料。 未来的阶层分化,从来不再取决于谁的脑子里背下了更多现成的公式, 而是取决于当 AI 把一万种完美方案同时拍在桌上时,谁拥有拍板做决定的判断力,谁能攒起一个活人团队去把事情干成, 所以别再逼着孩子在死记硬背的泥潭里跟硅基芯片比拼记忆容量了,当流水线式的教书匠和做题家同时迎来历史性失业,谁能先把教室变成训练人性直觉与创业野心的实战沙盘,谁才能在算力时代拿到唯一不被平替的入场券
372
1,565
243
8,170
850,292
Rahul Goma Phulore retweeted
i went to bed and asked claude to paint an autumn tree overnight. at around 4am, it realised that it didn't have some shades -- so it stopped and titrated its own paint: diluting red, sienna and madder step by step on a second palette, then measuring each pocket from a photo. woke up to this beautiful watercolor painting. 303 brush marks, took about 4.5 hours 🍂 (still working on the speed)
508
749
529
9,416
2,846,350
Rahul Goma Phulore retweeted
Generative UI for conversational agents. Combining @aisdk with Chat SDK and 𝚓𝚜𝚘𝚗-𝚛𝚎𝚗𝚍𝚎𝚛, you can go straight from LLM to UI on platforms like Slack and Teams. I believe this will become the dominant Human-Agent Interaction (HAI) paradigm: vercel.com/changelog/json-re…
257
145
55
2,841
221,316
Rahul Goma Phulore retweeted
we've completely rebuilt how AI agents work inside JetBrains IDEs it's called Air, and starting with 2026.3 it ships with all our IDEs Some features: ⛽︎ sign in with your existing ChatGPT, GitHub Copilot or Google account. Claude subscriptions work in terminal view ⛽︎ or bring your own OpenRouter, OpenAI, Anthropic or Gemini API key for supported chat agents ⛽︎ agents work right next to your code, so you can jump into any file they touch without leaving the IDE ⛽︎ hit a limit mid-task? switch to another subscription or API key right in the session, the work keeps going ⛽︎ Claude Agent, Codex, Junie and Copilot work out of the box. add Gemini CLI, Antigravity, OpenCode, Pi and more from the ACP registry, or connect your own agent through ACP ⛽︎ run agents in parallel in separate Git worktrees, then cherry-pick their changes back ⛽︎ agents get the IDE's symbol search, rename refactoring, inspections and run configurations, plus the debugger in IntelliJ IDEA and CLion ⛽︎ Junie Lite is free with a JetBrains account ⛽︎ leave notes on code or diffs and send them to your agent, Claude Code and Codex terminal sessions included ⛽︎ set up MCP servers once for every new agent chat, add skills from folders or GitHub ⛽︎ see at a glance which agents are running and which are waiting for you, use Caps Lock is LED of agents status ⛽︎ review the last turn, roll back individual files ⛽︎ stuck in a merge conflict? hand it to an agent right from the merge dialog ⛽︎ save agent presets and pin the prompts you keep reusing ⛽︎ macOS, Windows, Linux. Remote Development over SSH, WSL and dev containers too and yes, you can change the chat text size and much more features are coming before NY
31
20
16
138
31,206
Rahul Goma Phulore retweeted
Might be controversial, but: I have found regressions to be far less of a problem with agents than with humans.
46
12
7
367
17,994
Rahul Goma Phulore retweeted
Drag-and-drop makes the digital graspable, but building it well is fiddly enough that many don’t bother. Declarative Dragging (with Josh Horowitz & Jeff Heer, #uist2026 best paper award), lets you make an element draggable just by listing the states dragging it can reach.
46
225
44
1,779
110,143
Rahul Goma Phulore retweeted
maybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis
796
434
300
11,198
1,441,638