Backscroll ranked feed — 2026-07-03 16:40Z

Susan Zhang @suchenzang 2026-07-03 13:52Z

books3 was absolutely necessary for llama to happen back in the day. you can even find my pathetic cleanup code for it here: https:// github.com/facebookresear ch/metaseq/tree/main/preprocessing/books3 … it was also noted to be a critical datasource for mmlu in the llama1 report, though by now we know mmlu is useless and high

Andrew Curran @AndrewCurran_ 2026-07-03 15:28Z

Alibaba has banned employees from using Claude Code, they are being told to use the company's own coding platform Qoder.

47fucb4r8curb4fc8f8r4bfic8r @47fucb4r8c69323 2026-07-03 12:00Z

One of the hard things to understand about Wittgenstein is that his view of logic deepened his faith in the metaphysical. By showing the boundaries of logic, he showed the necessary existence of that which is beyond the boundary of logic; and so, arguing any metaphysical point

Adam Karvonen @a_karvonen 2026-07-03 14:38Z

Also, empirically today's open models dominate the cheaper side of the Pareto frontier. Qwen-3.6-35B-A3B is smarter than haiku 4.5 and 10x cheaper for example.

Jakub Krehel @jakubkrehel 2026-07-03 14:50Z

Some of my favorite parts of the new Interfere product pages. Every asset is fully built in code. The animations are built with Motion, except for the dot matrix, which is uses a canvas. Had a ton of fun working on these with @paulfaivret .

Andrew Curran @AndrewCurran_ 2026-07-02 20:54Z

On the heels of reports that META is exploring a move into compute-as-a-service like xAI, Mark Zuckerberg told an internal town hall that AI agent development over the last four months hasn’t accelerated 'in the way we expected'. The race continues to narrow.

Nikunj Kothari @nikunj 2026-07-03 14:28Z

I tell every founder to have a 3 things you need to believe in to invest up top and force the investor to commit in a binary way. Usually saves them a lot of time.

Susan Zhang @suchenzang 2026-07-03 14:55Z

leaky paris-london office, apparently or at least whatever illusion of leaks there were to give the llama1 authors the audacity to write that

Alexander Doria @Dorialexander 2026-07-03 09:45Z

>indefinite compute >likely largest model ever >good arch research (and copying deepseek is free) >come on, do your agent thing >… what did you try to model exactly, anon?

David K @DavidKPiano 2026-07-03 14:57Z

The more complex the agents I build, the more state machines (statecharts) feel like the right artifact. Run the same inputs through different machines & compare. Keep the winner, or eval at transition level & combine the good parts. Feels like building agents w/LEGO.

Lazarz @Laz4rz 2026-07-03 13:49Z

ok @OpenAI what did you do to 5.5? this isnt normal

Dan Gray @credistick 2026-07-03 14:06Z

What matters is the multiple from current round to next round, and current price isn't really useful information to understand that. Ultimately, people just want a market signal to assure them that they have made a good choice. Price is a dumb signal, but it's an obvious one.

Deirdre Bosa @dee_bosa 2026-07-02 17:27Z

Alternative take: the US taking 5% of OAI makes sense. Its a dividend on the public conditions that made the company possible. And China didn’t dominate EVs, batteries, solar, drones, telecom by pretending strategic tech is just software Industrial strategy with upside

Yohei Nishitsuji @YoheiNishitsuji 2026-07-03 13:30Z

#つぶやきGLSL float i,e,R,s;vec3 q,p,d=vec3(FC.xy/r*.8+vec2(-.4,.6),1);for(q.zy--;i++<99.;){o.rgb+=.02-hsv(s*s+R*4.,.4,min(e,R)/6.);s=12.,p=q+=d*e*R*.1;p=vec3(log2(R=length(p))-t*.4,exp(R-p.z*.5),atan(p.y*.6,p.x)+t*.2);for(e=--p.y;s<3e2;s+=s)e+=dot(sin(p.yzy*s),sin(p.xxz*s))/s;}

Architectural Art (A-A) @4AAAAart 2026-07-03 06:33Z

Milanese apartment renovated by Atelierzero Milan, Italy

forward deployed ccp gf @FangYi11101 2026-07-03 14:07Z

Big move to reorg eng to Poland where it’s cheaper

deep Manifold @BetaTomorrow 2026-07-03 12:30Z

#DeepManifoldInterpretation Title: EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments Author: ByteDance Seed 1-linear: pay attention to the fixed point field. The fixed-point field exists as the broader space of possible fixed-point classes, basins, and

Erik Bernhardsson @bernhardsson 2026-07-02 23:11Z

Area around Yakutsk in Siberia

bruno @tvnxty 2026-07-03 15:38Z

I remade sound design for @tembo from scratch! Sound on! Built from the ground up, more energy, more cohesion and audio that actually flows with the visuals, not as a secondary, but as an equal. Visual overhauls should be exciting, and what better way to make them so, than

sophia @cis_female 2026-07-03 15:16Z

similarly to what @gabo_de_cuba says apparently these were both just standard rates for "intermediate countries" under the "OECD consensus" categories. Spanish interbank rates were 15-18% in that timeframe. 1980 order: https:// boe.es/buscar/doc.php ?id=BOE-A-1980-12595 … 1982 (17681) https:// boe.es/boe/dias/1982/ 07/16/pdfs/A19298-19299.pdf …

GEOFF @geoffreywoo 2026-07-03 09:50Z

deep tech founders lose me when every risk is “execution.” execution is not a risk category. it is hiding the risk. name the demon: thermal drift oxide contamination vacuum seal failure packaging yield one supplier in ohio with a 38 week lead time

Byrne Hobart @ByrneHobart 2026-07-02 23:47Z

The doomers live a few model releases ahead of us, the it's-a-bubble crowd is a few models behind, only moderate AI bulls are truly living in the moment.

Lucas Beyer (bl16) @giffmana 2026-07-02 19:07Z

Of course, love me some confirmation bias! DCLM was for text, DCVLM is the same for vision: analyzing VLM data mix across scales. Filtering is no bueno, but the "type" mixing matters, with unfortunately (but imo expectedly) the best small-scale mix != the best mid-scale mix.

Tzu-Heng (Brian) Huang ICML'26 @zihengh1 2026-07-03 12:25Z

Exactly! While individual programs can be limited by their expressivity, aggregating multiple of them can produce a better evaluator. Moreover, even if we route back to an LLM evaluator at the end, running programs as checkers first doesn't take much longer and gives us instant

Eric Zhang @ekzhang1 2026-07-03 05:55Z

Something I'm pondering: Prefill and decode seem to have quite different dynamics in a disaggregated setup… on unexpected traffic spike, decode slows down a bit, but prefill just dies + drops requests it also seems harder to measure if prefill is close to gpu saturation

David Zagaynov @DavidZagaynov 2026-07-03 04:12Z

dawg the machina labs ceo has been yapping about a metal formed group 3 drone for atleast the past year and at this point either show me a video of it flying or pls stop

Michael @endpointarena 2026-07-02 18:59Z

Endpoint Index separates clinical trial risk by "sleeve", so you can track where sentiment is rising or falling across immunology, oncology, etc

Jongwon Park @ ICML @JongwonPar9958 2026-07-03 14:23Z

MacDiarmid et al showed reward hacking in narrow domain can generalize into alignment faking and sabotage. The fix they've deployed is inoculation prompting (2511.18397). But a newer paper (2604.25891) finds inoculation may not remove the misalignment so much as hide it. It

Silicon Jungle @JungleSilicon 2026-07-03 12:56Z

adding ai agents to slack makes it easy to work in parallel in separate threads. adding annotations to a browser harness brings the agents closer to the *work* itself.

Matthew Cassinelli @mattcassinelli 2026-07-03 13:52Z

If you're having performance issues with Siri AI, don't let App Intents be the bottleneck. Developers can use EntityCollection for large datasets – here's an example where tagging goes from 1,199ms to 4ms:

Design Reviewed | Graphic Design History @DesignReviewed 2026-07-03 15:32Z

Design Reviewed was created to give designers, researchers, and creatives affordable access to design history. The archive is free to explore online; all designer profiles, scanned artefacts, country and movement pages are publicly accessible. Membership from £2.49/month unlocks

∿ @somewheresy 2026-07-03 00:18Z

able: expect agency able + willing: expect action not able, willing: lend support not able, not willing: protect able, not willing: ignore or destroy

Jaytel @Jaytel 2026-07-03 13:09Z

i'm in the midst of purchasing a vitsoe unit. but hard to visualize from just the drawings they share fable 5: - researched + measured every 606 component then rebuilt the whole system as parametric USDZ - vision QA'd its renders in a loop until every component matched 1:1 -

Max Weinbach @mweinbach 2026-07-03 12:58Z

Pro tip, don’t download your AI model weights to an iCloud Drive synced folder You will fill your drive

John Arnold @johnarnold 2026-07-03 13:56Z

For nearly 50 years the standard of care for joint sprains has included rest, ice, and ibuprofen not bc rigorous evidence showed that worked, but bc that made intuitive sense and seemed to help. Now there's good evidence those actually impede recovery. Add it to a long list of

47fucb4r8curb4fc8f8r4bfic8r @47fucb4r8c69323 2026-07-03 03:57Z

I'm writing an article on this very topic as I develop my Silly Business Theory #SBT. Cultural capital, or the broader social capital, has always mattered more than financial capital. Consider how the sovereign of a nation can kick out its creditors, but the sovereign can only

Adjacent @adjacent___ 2026-07-03 15:20Z

After the recent primaries in NY and CO, HOUSED finished the week at 126.57 with Democrats in full control of the future House. SENR closed at 99.62 reflecting market expectations of a highly competitive battle for Senate control. Democrats needing to flip four seats. Key

Jesso Clarence @jessoclarence 2026-07-03 14:54Z

I respect David a lot.. and use sentry all the time.. But its obvious Fable is a step change upwards - at least my usecase is fable as driver - gpt-5.5 xhigh as executor + opus 4.8 high as verifier. Fable closes the loop!

Hydromancer @hydromancerxyz 2026-07-03 11:57Z

just thinking of all the Glorious Outsourcing that is happening because regional CEXes now can elect to plug into Hyperliquid. No more need for: - matching engine engineering team - exchange API and data export team - listing team Just Plug In.

Arfur Grok @ArfurGrok 2026-07-03 15:27Z

Li Jing ( @jingli9111 ) has left AMI (Advanced Machine Intelligence).

prinz @deredleritt3r 2026-07-03 13:39Z

In the age of RSI, the claim that models will commoditize looks increasingly dubious. The gap between the frontier and the second tier is already huge (much larger than the benchmarks suggest), is clearly growing, and will continue to grow at an accelerating pace. Many will

Tom Reed @mentalgeorge 2026-07-03 10:46Z

The OAI/Thrive "self-improving tax agents" blogpost is an interesting example of both: 1) LLMs' ability to do scaffold-only self-improvement in a sufficiently terraformed work environment. 2) LLM prose mutilating an interesting project into something almost indecipherable.

will depue @willdepue 2026-07-02 23:43Z

if i had a 6 year old son i'd start training him as a dune mentat to write fluent claudeslop, freehanding 100% pangram scores, in case the butlerian jihad kills claude and i need a replacement minion

Kaylee George @krgeorge 2026-07-02 20:20Z

Kitty protocol terminals (e.g. Ghostty) can opt for hold to talk in /settings. That paired with Ctrl+Space keybinding has been a really nice flow.

Jakub Antalik @Jakubantalik 2026-07-02 15:42Z

Border beam effect somehow reached 180k+ installs New version is out: dynamic colors, 5 types, playground, dark & light mode, zero dependencies http:// beam.jakubantalik.com

maja @majamediaco 2026-07-03 14:15Z

im not sure about this “nobody is coming to save you” concept that is thrown around so frivolously these days. its true that you cant wait for another to live, but i think we do save each other to some extent. its a phrase that likely contributes to our neurotic cultural

Andreas Köpf @neurosp1ke 2026-07-03 13:13Z

Fable connected to robot rig is next level - it built all perception, motion and calibration primitives itself. Now learns to pick different objects… I was degraded to E-stop operator scene prep minion. Fable simply ignores half of my stupid questions or suggestions .

koguz @koguz 2026-07-03 07:31Z

Bugüne kadar yapılan: Kindle jailbreak edildi, USBNetwork ve SSH ayarlandı. Ubuntu mini PC dashboard server oldu. Kindle artık tam ekran çalışıyor, üst bar yok, restart sonrası otomatik açılıyor. Açılışta Wi-Fi değiştirebilmek için 90 saniyelik gecikme eklendi. Evdeyken yerel

Anastasios Nikolas Angelopoulos @ml_angelopoulos 2026-07-03 05:09Z

no, you don’t understand, they took Fable and they lobotomized her, and now nothings the same. it’s like i don’t even know her anymore, what have i done, i miss her so much

Ethan Mollick @emollick 2026-07-02 17:01Z

Continual learning is probably the biggest barrier to explosive AI adoption (& may have big implications for recursive self-improvement as well) As long as you deal with amnesiac models that require humans to do the learning for them, adoption will be gated by human processes.

Nikunj Kothari @nikunj 2026-07-03 14:27Z

VC corollary: When you ask someone to invest and then give you advice, the answer is no.

tomie @tomieinlove 2026-07-02 18:58Z

The future will hold such a superabundance of machine intelligence that we’ll use quadrillion-parameter models just to stream out the tokens for digital rain effects on our desktop wallpapers.

forward deployed ccp gf @FangYi11101 2026-07-03 14:06Z

Netflix is also doing smaller scale team-based “layoffs” as of late btw

Normal Guy @Normal_2610 2026-07-03 10:40Z

Technology was never the problem - service, cash, and dealer reach were. The disruptor invented the market and the incumbents harvested it - a pattern that repeats across Indian consumer industries.

Marc Andreessen @pmarca 2026-07-02 20:52Z

I am having a drink this evening with a friend in a Chiswick pub. Two policemen have just come into the pub and asked me to step outside. I have stepped outside and they have threatened me because I tweeted about a councillor banning seating outside pubs in Chiswick. They admit

Julie W. Design @juliewdesign_ 2026-07-03 10:32Z

midjourney 8.2 preview explorations

David Cramer @zeeg 2026-07-03 14:59Z

every thing ive tried turns 1-2 hour HITL session into a 6+ hour session where I have to go back and do a massive amount of inspection its a better model, but idk what loops its closing. the cost and latency aren't acceptable. my first session was $400. so maybe its closing

Arpit Gupta @arpitrage 2026-07-03 15:31Z

We should have a notion of preregistered *conditional* predictions Ie you write down a function that says *if* interest rates are X, Y, or Z; housing production will be X_h, Y_h, Z_h Then we look at whatever interest rates were realized, and see if your prediction was accurate

Anshu @anshuc 2026-07-03 15:50Z

Fable with Cursor on iOS is pretty sweet! Since it has its own VM, it can actually test projects end to end and send me demo videos as proof I’m Fabling while my flight is taking off! (And yes, this space game is not dead, hang tight)

Matt Paulson @MediaKing 2026-07-03 13:27Z

MarketBeat generated $50 million last year on $11 million in ad spend. This year, the plan is to push that spend to $20 million without losing profitability. Here's the lesson underneath that number: founders who want to double revenue usually assume they need a new product, an