Adeel Siddiqui
48 posts
Adeel Siddiqui
@adeelsid7
data centers @landgatecorp • @ucla
Katılım Ocak 2023
231 Takip Edilen46 Takipçiler
Who read their bookmarks?
I find it super hard to read them in X.
I miss the time I built a tool for that, killed by X API prices 😢
English
@jeff_weinstein yeah it’s terrible. try landscape, makes some things easier
English
is editing and commenting on google docs on iphone absolutely garbage for everyone else, or is it just me somehow? (eg takes 2-10s to open/close comments etc.)
English
Now feels like a good time to share that I joined @GoogleDeepMind last month to work on compute.
I am very happy to be here and I can’t help but feel that I’m in the right place at the right time.
If you’re in SF let’s hang out.
English
Adeel Siddiqui retweetledi
Adeel Siddiqui retweetledi
Life Update: I moved to San Francisco.
Grateful to be here and inspired by the possibilities this city holds.
English
Adeel Siddiqui retweetledi
Blender MCP is here, allowing Claude to talk directly to Blender. For example, provide a 2D reference image and ask Claude to create it in 3D for you.
English
Adeel Siddiqui retweetledi
it's interesting that 1.5 billion parameters is all you need to crush math competitions, but you need like 15 trillion to make the model be funny
maybe humor is the right measure of true intelligence
English
cursor has been unusable for me.. "Failed to get complete AI response", no chat or cmd + k responses.. feels like im being rate limited or there is a bigger issue here?
I updated from older version and still no luck. Also tried using my own api keys and same results. Anyone else seeing this?
English
let non coders choose an interface like this and cursor can easily become the front runner in the vibe coding race
Riley Brown@rileybrown
I BUILT A CURSOR WRAPPER: USING CURSOR WITHOUT WRITING A SINGLE LINE OF CODE FULL VIDEO BELOW: I've been talking to hundreds of people who refuse to try cursor because of how scary the interface is. And fair enough. Terminal, Composer, Agent, Commands... It can be scary. This is why people are using other web based tools to create apps. Which is awesome! But after about 12 prompts, they seem to stop working effectively. Meanwhile, with Cursor you can create MASSIVE projects without coding perfectly fine, for days or weeks. So I got some ideas. I took it upon myself to build a new front end to Cursor, using Cursor without writing one line of code. 00:00 - INTRO 01:29 - LET'S START BY ASKING COMPOSER 15:01 - TESTING THE APP
English
long-content language modeling is still the toughest restraint to crack. DeepSeek’s NSA is a step in the right direction. this could be a big leap in scaling compute
check it out: arxiv.org/pdf/2502.11089
English
you can just order bodyguards
Nikita Bier@nikitabier
Over the last few months, I’ve been advising @bookprotectors: a new app for ordering an on-demand security detail. Or more simply: Uber with guns. Today, they’re debuting in Los Angeles and NYC at No. 3 on the App Store. If you have a hot date this weekend, pick her up in a Protector.
English
@jeff_weinstein I aspire to be as good at email as you are. We need a Jeff Email Guide
English
Has anyone built an MCP only (or MCP first) business yet? If so, email me. Or if you want to make one, email me.
=> jweinstein@stripe.com
English
another alternative that works fairly well is creating a "memory-bank .md" file and including in your cursor instructions aka sytem prompt that when you tell it to update its memory it should breakdown what has been done so far, what is currently implemented and working and how
English
This is why cursor struggles with larger codebases and more complicated setups as you use it for longer.
Embedded indexes are hurting cursor for power users. When code is not working properly but being indexed anyways.. this is bad. Try windsurf out or 👇
Kevin Hou@kevinhou22
We built Windsurf’s agent to not rely on embedding indexes like other tools out there. One size fits all retrieval simply doesn’t scale for monorepos. Instead, our agent uses tools that a human would to find the context is needs: - Grep / file find - File relation traversal (ie. AST parsing) - Web search + online docs - Parallel LLM-based searches (mquery, see video) Our agent will converge to human behavior but with infinitely faster speed and capacity to understand. youtu.be/DuZXbinJ4Uc?si…
English
Adeel Siddiqui retweetledi
Unfortunately, information on the internet will become increasingly closed in the coming years.
My prediction is some micro-transaction protocol will emerge:
1. User sets a max $ per (hard) question
2. An agentic LLM figures out which paid data sources to seek access to based on the budget and user priority
3. Access transaction and answer
LLMs will naturally get good at recommending $ budgets based on user questions. And a discovery market for these paywalled sources will probably also emerge.
Karri Saarinen@karrisaarinen
Interesting insight from @benthompson. In the world of Deep Research, secrecy might be useful strategy at times and private data sources more valuable. AI is blind without the sources. It doesn't have friends, relationships or conferences overhear things
English