Ainar (❖,❖) @ainarmathgeek

Math geek | WEB3 fan Joined January 2013

Tweets

47
Followers

112
Following

124
Likes

45

Sentient @SentientAGI

2 months ago

How does the GRID orchestrate intelligence? By routing, combining, and coordinating thousands of models, agents, and data streams, the GRID scales like a distributed brain. As it grows, its collective intelligence compounds, delivering toward AGI-level output. Here’s how the…

433 893 2K 116K 67

Download Video

Sentient @SentientAGI

2 months ago

Sentient’s mission is to ensure that Artificial General Intelligence is open-source and not controlled by any single entity. The GRID is on its way…

2K 35K 39K 1.1M 397

Download Video

Ainar (❖,❖) @ainarmathgeek

5 months ago

I pledge myself to Ritual! 🔥🔮🩸 @ritualfnd

0 0 0 11 0

Simon Willison @simonw

6 months ago

TIL that Claude Code increases the "thinking" budget to maximum if you tell it to "ultrathink", which is a step up from "megathink", which is a step up from "think" simonwillison.net/2025/Apr/19/cl… - via x.com/HamelHusain/st…

Hamel Husain @HamelHusain

6 months ago

12 37 478 218K 479

Download Image

35 91 1K 184K 771

Download Image

ℏεsam @Hesamation

6 months ago

a banger article on RL and LLM reasoning was just released by @rasbt. hands-down THE best intro that covers: > reasoning models basics > RLHF basics and PPO explanation > how DeepSeek-R1 reasoning models are trained > top papers about reasoning models > lessons we learned so far

6 138 948 52K 976

Download Image

Tom Dörr @tom_doerr

6 months ago

Open-source framework for building and running AI agents, connects to tools and APIs, no invite code needed

17 196 2K 151K 2K

Download Image

LangChain @LangChainAI

6 months ago

🔐🤖 Secure AI Assistant Learn to build production-ready AI assistants that securely interact with Gmail using LangGraph.js and Auth0. Features secure tool-calling, real-time streaming, and complete authentication integration. Explore the full tutorial in reply 👇

3 89 543 33K 473

Download Image

Sebastian Raschka @rasbt

6 months ago

Just shared a new article on "The State of Reinforcement Learning for LLM Reasoning"! If you are new to reinforcement learning, this article has a pretty generous intro section (PPO, GRPO, etc) I addition, I cover 15 recent articles focused on RL and Reasoning Models,

50 476 2K 196K 3K

Download Image

anton @abacaj

6 months ago

OpenAI and Anthropic are cooked. Google finally made the best model - not just in benchmarks but real use. I don’t know anyone in my circle using Claude anymore, a move I did not see coming

Paul Gauthier @paulgauthier

6 months ago

OpenAI and Anthropic are cooked. Google finally made the best model - not just in benchmarks but real use. I don’t know anyone in my circle using Claude anymore, a move I did not see coming

23 53 646 276K 147

Download Image

124 105 2K 245K 563

François Chollet @fchollet

6 months ago

Google quietly released a powerful recommender systems library optimized for JAX and TPUs, based on Keras. It's called RecML. It has native support for SparseCore (latest hardware for handling large distributed embeddings)

11 124 1K 68K 644

Download Image

Ashpreet Bedi @ashpreetbedi

6 months ago

If you want to build good Agents, you need to know how to stitch together models, tools, instructions, reasoning, knowledge, storage and memory. Here's how I use them (link to code included) 👇

15 76 610 95K 1K

Download Image

LangChain @LangChainAI

6 months ago

🔧🤖 MCP-Use Tools Just launched: An open-source library that connects any LLM to MCP tools for custom agents, featuring seamless integration with LangChain and support for web browsing, Airbnb search, and 3D modeling capabilities. Explore the implementation on GitHub 🚀…

14 230 928 63K 900

Download Image

Daniel San @dani_avila7

6 months ago

Introducing Auto Documentation powered by Codebase Knowledge Graphs With CodeGPT, we traverse your entire repo to build a Knowledge Graph that understands how your software works. Capturing relationships between classes, modules, and dependencies. This allows us to generate…

14 100 654 57K 975

Download Video

Andrej Karpathy @karpathy

7 months ago

I just vibe coded a whole iOS app in Swift (without having programmed in Swift before, though I learned some in the process) and now ~1 hour later it's actually running on my physical phone. It was so ez... I had my hand held through the entire process. Very cool.

578 1K 23K 2.3M 12K

Ainar (❖,❖) @ainarmathgeek

7 months ago

Is there is a chain, where AI agents can be trained on-chain - that's SVM L2s, especially Eclipse.

0 0 0 10 0

Alucard @Alucard_eth

7 months ago

WINNING ECOSYSTEMS ARE NOT BUILT BY CHANCE. @EclipseFND has every piece of the puzzle: ✅️ A technically ambitious roadmap with the GSVM (1M TPS). ✅️ A cohesive brand and mascot strategy (Turbo/Cow). ✅️ A clear economic vision with revenue sharing via App-Specific…

244 126 727 42K 15

Eclipse Labs @Labs_Eclipse

7 months ago

GSVM In this article the Eclipse Research Team introduces GigaCompute and explains how we're using it to build GSVM (GigaCompute SVM): our faster, more efficient Solana client specifically designed for the L2 context. eclipselabs.io/blogs/why-perf…

83 152 501 28K 13

Download Image

Eclipse Labs @Labs_Eclipse

7 months ago

1/ The Eclipse Performance Thesis Eclipse is building the Giga Scale Virtual Machine (GSVM), a new SVM client. A software-hardware co-designed, cross-layer optimized, GigaCompute blockchain that eclipses all scalability limits and represents a paradigm shift in performance.

364 614 2K 778K 175

Download Image

Ashpreet Bedi @ashpreetbedi

7 months ago

> open source > multimodal > lightning fast > 80 toolkits with 1000s of tools > 23 model providers > 12 vector databases Built from scratch, 16 hours a day, 7 days a week.

53 200 2K 254K 3K

Download Image

Andrej Karpathy @karpathy

7 months ago

@simonw Good post! It will take some time to settle on definitions. Personally I use "vibe coding" when I feel like this dog. My iOS app last night being a good example. But I find that in practice I rarely go full out vibe coding, and more often I still look at the code, I add…