unicorn🦄🇨🇮🤹🏾‍♂️

🧠 🗣️Language (speech) Technologies | Deep Learning Engineer | He/him | Luke 12:48 | let’s tell the world our own story 😊

[Sebastian Raschka @rasbt]

Interesting surprise drop from Thinky! The Inkling model looks pretty solid on benchmarks, and it has some little surprises in its architecture:

- Small conv layers in several places
- An RMSNorm for the embeddings (before the block RMSNorm)
- Rel. position bias instead of RoPE

[unicorn🦄🇨🇮🤹🏾‍♂️ @cmDrStrange]

@SimonTheNoob @IntCyberDigest Come on, Claude 😅

[unicorn🦄🇨🇮🤹🏾‍♂️ @cmDrStrange]

Share your insights here (no personal data recorded)

Survey: https://t.co/buOAPDRY8v

Help us change the narrative and feel free to share this with your technical networks!

#DeepTech #ArtificialIntelligence #AfricaTech #BrainDrain #ComputeInfrastructure #TechTalent

[Doudou BA 🦑 @doudou_octopus]

I literally burnt my last $400 in a flight to Paris 🇫🇷 to attend a 36h YC / Paris Build Hackathon and get a chance to win a @ycombinator interview…
I don’t have a return flight ticket so if you see me in CDG Airport or Paris Metro you’re welcome to drop a coin… this could be your angel investment… seriously… not homeless … just a founder building OctOpus… will be a unicorn: )
@dessaigne

[NVIDIA @nvidia]

Water usage has been a hot topic in the AI data center world, but the numbers may surprise you.

According to the Manhattan Institute, data centers use 0.2 percent of daily water usage in the U.S. and that number has dramatically decreased in the past few years due to a new method: liquid cooling.

By moving to 45°C liquid cooling, AI factories in favorable climates can use dry coolers instead of conventional cooling-tower-based systems, cutting facility cooling water use from roughly 2.6M gallons per MW per year to near zero.

Liquid cooling enables AI factories to be both water and energy efficient, while creating opportunities for heat reuse and dispersal to local communities, allowing these factories to become energy grid assets.

Learn more below ⬇️

https://t.co/7WanoPNKTR

[Anatoli Kopadze @AnatoliKopadze]

NVIDIA CEO, Jensen Huang:

"Nobody writes prompts anymore. The new job is to write and handle loops."

This is the shift that's going to define the rest of 2026.

53 minutes of pure insight from one of the richest men on earth.

Watch it, then read the full guide on how to actually use loops below.

[Matthew Berman @MatthewBerman]

OpenAI might swallow up every piece of software eventually.

[Andrew Ng @AndrewYNg]

One of the new, buzzy jobs in Silicon Valley is the AI Forward Deployed Engineer (FDE), an engineer who is embedded within a client organization to help customize solutions, such as building and tuning agentic workflows that suit the client’s particular needs. I’ve heard from people who are wondering anew about the FDE career path since OpenAI and Anthropic started building new teams to place FDEs within client organizations.

The rise of FDEs for AI workloads is one way AI is creating new jobs (and why the jobpolcalypse narrative of upcoming job market collapse is false -- there will be many AI and non-AI jobs). However, I believe there will be far more AI Engineer jobs than FDEs, as I explain below.

The FDE role was pioneered about two decades ago by Palantir, which sent engineers to government locations to work on secure, air-gapped networks. In addition to having good technical skills, FDEs need communication skills and sometimes business skills. For example, they may need to speak with clients to understand their needs, formulate a strategy to prioritize projects, explain complex technology, and respectfully push back if a client asks for something unrealistic. They’re enjoying a resurgence because of the amount of work involved in taking an off-the-shelf LLM and building it into a custom agentic workflow that fits particular business needs.

However, I believe the number of AI Engineer jobs will be far larger. A company might accept a few FDEs to be embedded within its organization. But most companies will want far more of their own employees working on their projects. While my organizations do hire FDEs, we hire far more AI Engineers! Also, a common client concern is that it is hard to find vendor-neutral FDEs — they are, after all, there to deeply integrate a particular vendor’s product into a company. In this moment when it’s hard to predict which AI service will be the best one in a year’s time, optionality (the ability to pick whatever vendor turns out to fit best in the future) is very valuable. In contrast, letting FDEs tightly bind a company’s processes significantly reduces optionality.

Right now, I see surging demand for AI Engineers who can build software applications using AI software components (like LLM prompting, agentic frameworks, evals, etc.) and effectively use AI coding agents (like Claude Code, Codex, Antigravity CLI, and OpenCode). As the AI Engineer role matures, I expect it to fragment into more specialized roles, like the generic Software Engineer role from decades ago fragmented into frontend, backend, mobile, data engineering, devops, and so on.

What will be the future, specialized AI engineering roles? I don’t know. Perhaps there will be AI FDEs, LLMOps Engineers, Evals Engineers, AI Data Engineers, Harness Engineers, and other roles we don’t have names for yet. But for now, I see a lot of AI engineers who are generalists create a lot of value. Skilled AI Engineers are in very high demand! As our field continues to mature over the coming decade, I look forward to new specializations within AI Engineering that create even more job opportunities.

[Original text: The Batch newsletter]

[Andrej Karpathy @karpathy]

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

[unicorn🦄🇨🇮🤹🏾‍♂️ @cmDrStrange]

@LucaVaran Yako maman

[Sebastian Raschka @rasbt]

New article: a visual tour of recent LLM architecture advances, from Gemma 4 to DeepSeek V4.

I focus on long-context efficiency tweaks like KV sharing, per-layer embeddings, layer-wise attention budgets, compressed attention, and mHC.

Link: https://t.co/KO81y3kTH7

[unicorn🦄🇨🇮🤹🏾‍♂️ @cmDrStrange]

Elle parle s’amuser ainsi avec la même femme noire qui trouve l’homme attentionné faible, mou, collant, pas virils 🤔?

[Varan 🐊 2.0🇨🇮 @LucaVaran]

Elle a raison

[Noah @NoahKingJr]

Anthropic with Software engineers:

[Doudou BA 🦑 @doudou_octopus]

https://t.co/AVY6SKT2lI

[vas @vasuman]

12M context
52x faster
95% cheaper

What

[Ryan Peterman @ryanlpeterman]

Barbara Liskov (Turing Award Winner): "Python has modules, but it doesn't have encapsulation.

It allows code on the outside to muck around with what's going on on the inside of a module. Encapsulation is a crucial part of making modularity work.

And when you're building big programs so you have many programmers working on them, your team is really only as strong as your weakest programmer.

So it's nice if the compiler can enforce things and make certain kinds of bad behavior not possible."

[Matthew Berman @MatthewBerman]

Demis says he wants to see a Western open source AI stack and that we’re losing to China.

He also says Google doesn’t have enough compute to build two frontier (open and closed) models, which is why Gemma is a smaller family of models.

Watch this incredible clip.

Shout out @ycombinator and @garrytan for the fantastic interview.

[Sara Hooker @sarahookr]

Not specific to opus, but consistent with other recent scaling efforts in model size (gpt 4.5 last year)

Rates of return for compute are much better placed in post training, alignment, data targeting, gradient free learning.

It is the slow death of brute force scaling alone. innovation now lies in how a model interacts with the world.