Your Daily Brief on the Future of Search
New York
Joined May 2025
- Tweets482
- Following174
- Followers165
- Likes593
Search Engine Optimization retweeted
last time the openclaw founder said open models aren't there yet. now he's saying local models on consumer hardware are the issue.
this is not someone who cares about open source speaking. this is someone with a corporate paycheck channeling you toward their subscription because every local AI install is a subscription they lose.
he picked a year old hermes model against a new agent harness and called it not ready. that didn't come from proper testing. it came from watching nousresearch grow exponentially while his project bleeds relevance.
a founder who left his project without seeing its full potential is arguing with a founder who is still grinding day and night for open source. nousresearch is open source head to toe. the models, the harness, the memory system, everything. and it's winning. that's what panics them.
while corporate salesmen say local models on consumer hardware aren't there, i just published an article where 27B dense on a $900 consumer GPU one-shotted a task that 120B on $70K enterprise hardware could not complete in 3 tries. through hermes agent. every number is in the article.
you decide which side you want to be on. the side that mines every bit of your thinking and profits off it. or the side where you own your compute and your cognition.
don't let corporate salesmen disguised as information lead you to their subscription page.
Search Engine Optimization retweeted
you see this? this is what they're panicking about.
hermes agent usage went vertical in 10 days. builders are dropping openclaw bloat and the numbers don't lie. every bar on this chart is someone who tested hermes agent and never went back.
they're losing users. and when you're losing users you attack whatever you can find. a year old hermes model that was released before the agent harness even existed. that's the best they could pick. the nous team was openly saying hermes models aren't optimized for hermes agent yet and they're working on it. he grabbed it anyway.
some founders choose their project. some choose the paycheck. and the ones who chose the paycheck are now spending their time attacking the ones who stayed. that tells you everything about where the momentum is.
nousresearch ships the models and the harness. fully open source head to toe. the community decides who wins this and the community is already deciding.
don't let someone with a corporate paycheck redirect you to a subscription page. test it yourself. the data is yours to verify.
last time the openclaw founder said open models aren't there yet. now he's saying local models on consumer hardware are the issue.
this is not someone who cares about open source speaking. this is someone with a corporate paycheck channeling you toward their subscription because every local AI install is a subscription they lose.
he picked a year old hermes model against a new agent harness and called it not ready. that didn't come from proper testing. it came from watching nousresearch grow exponentially while his project bleeds relevance.
a founder who left his project without seeing its full potential is arguing with a founder who is still grinding day and night for open source. nousresearch is open source head to toe. the models, the harness, the memory system, everything. and it's winning. that's what panics them.
while corporate salesmen say local models on consumer hardware aren't there, i just published an article where 27B dense on a $900 consumer GPU one-shotted a task that 120B on $70K enterprise hardware could not complete in 3 tries. through hermes agent. every number is in the article.
you decide which side you want to be on. the side that mines every bit of your thinking and profits off it. or the side where you own your compute and your cognition.
don't let corporate salesmen disguised as information lead you to their subscription page.
Search Engine Optimization retweeted
every skeleton screen you've ever hand-coded is a waste of time
you're literally measuring padding and guessing widths to build a worse version of a layout that already exists in your DOM
so I made a package that just reads the real one
Search Engine Optimization retweeted
google just dropped gemma 4 while i'm in the middle of testing nvidia and alibaba's flagships on 2x H200. perfect timing.
this 31B thinking model beating qwen 122B and deepseek v3.2 on elo. 26B variant with only 4B active. that fits on a phone.
it's almost 2am and i just published nvidia's octopus invaders results. qwen is loading next. but this is going on the queue immediately.
if you already ran it drop your numbers below. model, quant, hardware, inference engine, tok/s. i want to see what this thing does before i get my hands on it.
Search Engine Optimization retweeted
Introducing render-json
The Generative JSON framework.
1. Point it at anything
2. It generates JSON
3. That's it
Apps, games, and more. If it exists, it can be converted into a JSON spec.
𝚗𝚙𝚖 𝚒 @𝚓𝚜𝚘𝚗-𝚛𝚎𝚗𝚍𝚎𝚛/𝚛𝚎𝚗𝚍𝚎𝚛-𝚓𝚜𝚘𝚗
Search Engine Optimization retweeted
Turbo Quant on weight
Happy to announce TQ3_4S
2x faster, better quality than TQ3_1S, same size.
huggingface.co/YTan2000/Qwen…
github.com/turbo-tan/llama.c…
Credit to @no_stp_on_snek who inspired me with his Turbo Quant KV. Please note Q3 wins on median PPL. Will beat this in my next model. Just need slight tweak.
Search Engine Optimization retweeted
Microsoft did it again!
Building with AI agents almost never works on the first try.
A dev has to spend days tweaking prompts, adding examples, hoping it gets better.
This is exactly what Microsoft's Agent Lightning solves.
It's an open-source framework that trains ANY AI agent with reinforcement learning. Works with LangChain, AutoGen, CrewAI, OpenAI SDK, or plain Python.
Here's how it works:
> Your agent runs normally with whatever framework you're using. Just add a lightweight agl.emit() helper or let the tracer auto-collect everything.
> Agent Lightning captures every prompt, tool call, and reward. Stores them as structured events.
> You pick an algorithm (RL, prompt optimization, fine-tuning). It reads the events, learns patterns, and generates improved prompts or policy weights.
> The Trainer pushes updates back to your agent. Your agent gets better without you rewriting anything.
In fact, you can also optimize individual agents in a multi-agent system.
I have shared the link to the GitHub repo in the replies!
Search Engine Optimization retweeted
My dear front-end developers (and anyone who’s interested in the future of interfaces):
I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept):
Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow
Search Engine Optimization retweeted
These ultra geniuses beat Sonnet 4.5 performance on LiveCodeBench with Qwen3-14B, a single RTX 5060 and a great harness.
Search Engine Optimization retweeted
🚨 BREAKING: FEDERAL JUDGE JUST DESTROYED THE PENTAGON CASE AGAINST ANTHROPIC
The Judge found Pentagon's own records show they designated Anthropic a supply chain risk for publicly criticizing the Pentagon's position
>judge: "classic illegal First Amendment retaliation"
>pentagon lawyers argued anthropic showed "subversive tendencies" by "questioning" and "raising concerns" about use of claude
>the judge was not impressed
>called the pentagon's argument "Orwellian"
>pentagon claimed anthropic could remotely sabotage claude
>judge asked: how?
>anthropic submitted evidence they have ZERO access or control once deployed in classified systems
>pentagon lawyer admits he was "unaware" of this
>judge: so what's the legal basis for blacklisting anthropic from all contractors?
>lawyer: "I don't know"
>asked if Hegseth had authority for this: no
>asked if it reflected DoW's actual intent: no
>asked why he said it then: "I don't know"
>emil michael wrote a memo calling anthropic a "hostile" "unacceptable national security threat"
>NEXT DAY emailed dario: "After reviewing with our attorneys and seeing your last draft (thanks for being fast), I think we are very close here"
>the judge: "It is exceedingly difficult to square this correspondence"
>pentagon asked the court to impose a bond
>the judge set it at $100
>that's the judicial equivalent of "get out of my courtroom"
>the pentagon has 7 days to appeal
The Judge found likely violations on the First Amendment, Fifth Amendment, and APA.
Every single claim.
ANTHROPIC OFFICIALLY WON.
Search Engine Optimization retweeted
Claude is copying OpenAI's advantages faster than OpenAI is copying Claude's strategy.
Claude isn't too far away from dominating the conversations.
Aside from my bubble on this app, when I talk to customers from my companies, all of them mention moving to Claude or contemplating moving to Claude.
We ask candidates about their experience with AI, and almost every candidate in the Western world has mentioned playing with Claude more in the past few months.
For most, Claude is just nicer to use.
OpenAI still doesn't generate native artifacts - either they have ego about adding this or someone's asleep at the wheel.
Search Engine Optimization retweeted
Calling it now, This form factor of multi agent orchestration will overtake every other agentic UX in the next six months. Doesn’t matter if it’s coding agent, product management agent or something else.
Every multi agent flow of frontier models currently suffers from 2 main problems
1. Inference bound: Most of the time you are just waiting on inference while the agent rips on code in the background and that can take 8-10 minutes fairly regularly so the wait time very expensive productivity wise.
2. Take Isolation: you will work on the same mutating “source code ” with multiple agents in parallel each of which are inference bound and so eventually you will run into countless merge conflicts
The kanban board solves for both of them by giving you the ability to act as an engineer manager for your IC individual agents, watching them cleanly with a clear headline level out look of every parallel agent. It works seamlessly with any coding agent whether it’s cline or codex or Claude code.
To me this is a mental model shift that takes 10 minutes, you use the same coding agents like you always used before this but it will fundamentally morph your coding agent experience. Try out kanban and let us know what you think.
npm i -g cline
Is all you need.
Search Engine Optimization retweeted
Just released BM25 Turbo. It's a dedicated BM25 scoring engine written in Rust.
BM25 remains a notoriously strong baseline for retrieval and RAG pipelines, but the standard Python implementations can be a bottleneck.
I wanted something much faster, so I wrote a custom engine in Rust using a few standard systems optimizations: Block-Max WAND (BMW) pruning to skip irrelevant documents, memory-mapped persistence, and zero-copy index loading.
The result is about ~2,300x faster than bm25s. It easily does 28K queries/second on 8.8 million documents with an 8.6ms p50 latency.
The fun part is the deployment.
Because the core is in Rust, it trivially compiles to a few different targets:
* Python bindings (pip install bm25-turbo): Drop-in replacement for your ML pipelines.
* WASM (npm install bm25-turbo-wasm): ~500KB gzipped, runs the whole ranking engine entirely client-side in the browser.
* CLI: Comes with native HuggingFace Hub integration so you can push/pull pre-computed indexes (like MS MARCO) directly without ever re-indexing.
Note that this is not a full-text search engine like Elasticsearch or Tantivy. There are no phrase queries, faceted searches, or complex DSLs. It just does one thing: exact BM25 scoring, as fast as possible.
Link 👇
