In software engineering, lines of code edited and weekly merge counts are misleading proxies for productivity. There are a significant number exogenous variables that impact those metrics. To name only a few - team dynamics, code maturity, product maturity, business seasonal cycles and many more.

In this age of AI, don’t be a follower. Be the leader who hires engineers who build the future - because AI ain’t building jackshit for you.

We are entering the golden age of AI “world models” where every AI hype will be proudly accompanied by their grand unified theory of everything, rigorously engineered to collapse at the first gentle poke of reality.

Classifier ≠ Human Judge > We assess how effectively large language models generate social media replies that remain indistinguishable from human-authored content when evaluated by automated classifiers. We employ a BERT-based binary classification model to distinguish between the two text types.

AI isn’t going to wound web’s ad model - fatally or otherwise. AI companies are going to be the ones serving those ads. I would be shocked if OpenAI hasn’t / isn’t already indexing the web even as I type this.

Demo coming soon … bsky.app/profile/avik...

I have said this before over on the bird site (don’t want to visit to find it - it’s extra toxic right now given today’s election results), but many of the young folks in “AI” seem to suffer from the affliction of never having worked with good old ML, before they got mesmerized by “AI”.

Every author writing like this should be required to rewrite abstracts in plain English and read it aloud to an audience of their peers, before they can publish it. Summary: Conjectural with nice diagrams but no quantitative measures and ignores prior literature. arxiv.org/pdf/2510.26745

Unfortunately, at this point, any ‘AxI‘ naming is tainted. Whether we use “General”, “Super”, “Hyper” or [Insert], it’s an academic distinction without real world difference. That it also attempts to name a class of models that don’t actually exist, reinforces the conjuring of hype over substance.

LLMs have no utility - is not something I can subscribe to. LLMs have low utility relative to investment - is my stance. Utility is tied to cost. Cut LLM spend by 1000x with SSLMs and value equation shifts. Smaller cheaper task-tuned models FTW. Yes, you still have to pay the humans. Sorry?

Karpathy’s tweet is a live demo of the learning loop he promotes. Consciously or not, he is channeling: - Kolb: Experimental learning theory - Feynman: Explain in your own words - Dweck: Growth mindset scale The medium is the message.

In the research for Computing, my multi-part documentary that examines the intersection of computing and what it means to be human, I've collected almost 6,000 books to help inform my storytelling. You can browse my entire collection here t.co/fw6RXUYR2l

Don’t have an exact number, but 150+ trick or treaters tonight. One of them: K (Kid): Trick or treat? M (Me): Trick. K: Huh? M: What’s the trick? K: You give me candy. She was the youngest one of the evening. Cutie pie at her best!

Finally … www.wsj.com/tech/ai/larg...

In their own words: “Several caveats should be noted: The abilities we observe are highly unreliable; failures of introspection remain the norm.” transformer-circuits.pub/2025/introsp...

You know how they are going to react to this? Layoff another 100k human engineers to build another super data center because they are convinced they are on the brink of breakthrough - just needs a bit more juice. Newsflash, boys: AI broke thru back in 2023, now you are just chasing the ghost of AI.

When did making kids go hungry become a Christian value?

Open source xLMs are rapidly closing the gap with foundation models specially for custom tasks. Stop wiring directly to vendor APIs. Yes, it might delay your product launch by a few months and that maybe a deal breaker for startups. But if you are an enterprise, you should have no excuse.

On Bsky when posting about scientific papers or articles - I have two modes: 1. Has substance? Put on my scholar hat, assess it in its own register and respond with rigor. 2. Is basic? Hat stays off and I keep it casual because it’s not worth the time. On “AI“ these days, most fall into the 2nd.

Wait till he gets to the touch problem.

Root cause was lack of proper versioning and concurrency control in distributed configuration management system which violated ordering guarantees resulting in stale state propagation with cascading service disruption. Frankly, shocked that it hadnt happened before. aws.amazon.com/message/1019...

Soon, AI will also deliver a baby in one month with nine women: www.wsj.com/tech/ai/ai-r...

For an author of a transformative paper it takes a lot to say this, but, it’s been apparent for a while: > "Despite the fact that there's never been so much interest and resources and money and talent, this has somehow caused the narrowing of the research that we're doing," Jones told the audience.

If you have ever worked with any ML model, this outcome has always been extremely predictable. arstechnica.com/ai/2025/10/r...

If even Karpathy can’t get AI coding to work for him, are you willing to bet on it working for you? Your’s are IID you say? You will soon find out dimensionality means yours are OOD too.

Yeah, it should have been way more discussed than it was: bsky.app/profile/avik...

Could someone please tell the Amazon app that the AWS outages have been resolved couple of hours ago? Thank you.

Guess my AWS delivery is not happening today.

AWS Outage RCA: Jeff wants to announce 50% of our code is AI generated before Satya can.

Today for some reason, I felt the urge to share this tweet posted back in 2018:

This AWS outage is going to hurt startups and SMBs the most. If you are an enterprise you have redundancies in place, well - let’s say you should have had them in place, but startups and SMBs often can’t afford that because they are running on such thin margins, if any.

Observation about Bluesky from the ongoing AWS outage - it takes an inordinate amount of time for topics to trend on Bluesky. I understand why that might be, but there are better models to assess emerging trends than just raw velocity and volume.

So, no break fast? Seriously though this is sad even as it explains some of the behavioral patterns visible amongst the folks most likely to have had these meal substitutes.

GPT-5 is actually better than GPT-4o if you pick the right KPI - cost efficiency. Yes, if you focus on model efficiency - it’s worse. But, with their unsustainable run rate - which one do you think OpenAI would choose to focus?

Read this if you work with LLMs. For those of us who have been hands on, both under the hood and in front of the shiny bits, it’s been obvious from the get go. Folks still refusing to acknowledge the obvious are invested in it, directly or indirectly. For the rest of us it’s just another tool.

Actual researchers are spending so much time fending off gAI hype it seems to be taking away from actual research to build the next version of NNs. Instead of trying to put guardrails, templates, tools and retrieval to fix what is an intrinsic problem, industry needs to pivot to better research.

Expecting LLMs to stop hallucinating because you gave it prompts, tools and external retrieval is highly optimistic. It will also “discover” new pathways for hallucination with any templates in the context window. Old hallucinations will just be replaced with new ones. www.anthropic.com/news/skills

Vibe coding company saying vibe coders are destroying their business model, is … a vibe.

One year of deaths and destruction could have been avoided. Frankly it’s also shocking that in these moments there is a trend of Democrat presidents taking the “high” road instead of following the “just” road. They should have gone public on each of these occasions but they chose not to.

Here’s the more interesting part: > The IMF is forecasting a smaller decline in U.S. consumer price inflation for 2025 to 2.7%, declining only to 2.4% in 2026, Gourinchas said. A year ago, the IMF had forecast that U.S. inflation would be back to the Federal Reserve's 2% target level this year.

If even Karpathy can’t get AI coding to work for him, are you willing to bet on it working for you? Your’s are IID you say? You will soon find out dimensionality means yours are OOD too.

Reminds me of the current “AI” environment … not sure why … “You can also be a bulwark of civility against the quicksand of approximation and post-truth.” apnews.com/article/pope...

When reading the next “advance” in LLMs discern whether it optimizes efficiency or improves efficacy. Efficiency gains are easier and follow efficacy gains because functional correctness trumps optimized performance. Vendors are focused on efficiency because models have hit their saturation limits.

Nice paper. Observations: - Verifier assisted code snippet optimization - Before solutions are sub-optimal as are the after solutions - Some of the paper’s commentary is contradictory - If framework included chaos monkey to simulate real world, would these hold? arxiv.org/abs/2510.061...

> Experts have become increasingly skeptical of the industry’s claims, though, as it becomes clear that AI isn’t yet ready to replace workers. Industry isn’t dumb enough to claim this, a handful of CEOs are.

Without neural plasticity there is no human intelligence and no elastic weights and replay buffers won’t solve the problem because the old weights are “forgotten” instead of co-adapting the new weights relative to the old weights.

Hoped this was true but does not seem to be. Bsky uses Hive for content moderation. Hive’s ToU clause 5.3 clearly states: > license to reproduce, distribute, publicly display and perform, prepare derivative works of, incorporate into other works, and otherwise use and exploit

Reason why Sora and other video, image and writing is getting wider acceptance is because artists art evaluation is a subjective process. Well most artists, good coders are artists too but the evaluation process is via objective functions, which LLMs are mathematically not capable of guaranteeing.

If this is the response from a CEO and “steward of AT Protocol”, whose site was languishing till most of us decided to posters’ strike her competitor, not sure building on that protocol is a good investment for any community - devs, mods or users.