😺 How to train your robot

Welcome, humans.

Poll time! We’re working on some longer form pieces that we think will be really worth your time, but we need some help figuring out what to prioritize first.

Which of the below topics interest you THE MOST? Pick your top choice, as in the thing you want to learn ASAP.

What do you want to learn about AI?

Pick your favorite, then share any others in the "additional feedback"

Here’s what you need to know about AI today:

• HuggingFace released Mini Reachy, a DIY open-source robot.
• AWS and Anthropic announced an agent store.
• Robinhood’s CEO goes all in on new AI to solve hard math problems.
• Research showed AI use slows down devs instead of speed them up.

Advertise in The Neuron here.

Meet Reachy Mini, the tiny new open-source robot leading the DIY robot revolution…

Hugging Face, the central hub for open AI projects, just released a $299 robot that’s completely open source—and it might be your new desk buddy. We wrote a deep dive on Reachy and how this little guy represented the new face of the open-source robot revolution.

Your open-source companion - Reachy Mini

Remember when robots were either million-dollar factory arms or creepy Boston Dynamics videos that made you question humanity's future? Well, Hugging Face and Pollen Robotics just released Reachy Mini, a desktop robot that costs less than most people's monthly grocery bill, and might be the most adorable little robot we’ve ever seen

Here's the deal: Reachy Mini is an 11-inch tall, open-source robot that you can program in Python right out of the box. Think of it as the friendly cousin of those intimidating industrial robots, but one that actually wants to hang out on your desk and maybe help with your coding projects.

You can get two versions:

• The Lite version ($299), which needs to be plugged into your computer.
• The full version ($449), which has its own Raspberry Pi brain, WiFi, and battery so it can roam free (well, as free as an 11-inch desk-bot can be).

What makes this little guy special:

• Expressive personality: Motorized head that can look around, rotating body, and animated antennas that react to conversations.
• AI-powered senses: Built-in camera, microphones, and speakers for visual and audio interactions.
• Plug-and-play behaviors: Ships with 15+ pre-built robot behaviors you can download from Hugging Face.
• Community-driven: Upload and share new robot behaviors with Hugging Face's 10M+ users.
• DIY assembly: Comes as a kit you build yourself (great weekend project with kids).

The coolest part? It integrates directly with Hugging Face's ecosystem, meaning you can tap into state-of-the-art speech, vision, and personality models.

Want your robot to have conversations? There's a model for that. Want it to recognize objects? Yep, model for that too.

Why this matters: We're entering the era where AI is becoming physically embodied. While everyone's been focused on chatbots and image generators, companies like Hugging Face are quietly building the infrastructure for AI that can actually interact with the real world.

At $299, Reachy Mini isn't just democratizing robotics; it's creating a new generation of AI builders who will grow up thinking “of course my robot can understand what I'm saying.” These desktop robots could be the training wheels for the next wave of AI applications.

The bigger picture: This feels like the Raspberry Pi moment for robotics. Just like how that $35 computer sparked a maker revolution, Reachy Mini could be the gateway drug that gets thousands of developers, students, and curious tinkerers into physical AI.

Plus, it's open source everything—hardware designs, software, even simulation environments. That means the community can improve and expand on it faster than any single company could.

Worth checking out: You can order now with delivery starting late summer 2025, or dive into the Hugging Face community to see what people are already building in simulation. Read more about Reachy and the open-source revolution here!

Prompt Tip of the Day

Stop wasting time re-prompting the same tasks over and over. Here's the hack that changed our workflow:

The first AI response to your prompt is usually the smartest (the longer the chat, the harder it is for the AI to “focus”), but iteration helps you provide feedback to the AI to get what you truly want.

After you've iterated your way to the perfect AI output, ask the AI: “Now, write me a single prompt that would provide the exact same output in one shot.”

Then, save THAT prompt in as the custom instructions in a project for instant reuse (name the project after the task the prompt accomplishes).

This turns your projects into a personal prompt library, while also teaching you better prompting skills (“ohhh, so THAT’S what I should have said).

Check out all of our Prompt Tips of the Day from June here.

P.S: Completely new to AI? Start here!

ICYMI: Our latest podcast episode!

In this episode: Corey and Grant chat with AI implementation expert Andreas Welsch on whether AI will lead to a productivity boom (and how to apply it in your business) or lead to an AI slop-filled bust (spoiler: it’s all about how you use it!).

Listen here: YouTube | Spotify | Apple Podcasts

P.S: Want to reach 570,000+ AI professionals through podcast (or newsletter) ads? Fill out this quick form. We’d love to work with you!

Treats To Try.

*Asterisk = from our partners. Advertise in The Neuron here.

1. *Murf AI turns your text into human-like speech in 200+ voices, so you can narrate presentations or create audiobooks without recording anything yourself.
2. FlexOlmo from the Allen Institute (the actual open-source AI org) lets you contribute your sensitive data (like medical records or financial documents) to train language models without sharing the actual files, and you control when your data contribution is active and can turn it off anytime (paper).
3. Ikiform is a cheaper version of Typeform that builds designed forms (like contact forms or surveys) for you instantly just by describing what you need, with unlimited responses and analytics (here’s an open-source alt, from the makers of FormShare)—read more.
4. Knox speeds up your federal security compliance (FedRAMP) from 3 years to 3 months so you can sell software to government agencies like the Air Force (raised $6.5M).
5. LGND is “ChatGPT for the earth”; it turns satellite images into insights for tracking deforestation, crop disease, and pollution at 99% lower cost.
6. Google Gemini now lets you generate videos (via Veo 3) from images inside the app (read more).
7. Speaking of AI browsers, BrowserOS runs AI agents locally in your browser to automate tasks while keeping your data private, offering a Chrome alternative with built-in automation—open source, so free.
8. Check out this list of infrastructure companies who have been accepted as part of Google’s “AI American Infrastructure Academy” (lots of cool upstarts!).

See our top 51 AI Tools for Business here!

Around the Horn.

• It turns out, many “missionaries” (to quote Sam Altman) turned down Meta’s “mercenary” $100M signing bonus offers after all…but more of them stayed at Anthropic and DeepMind than at OpenAI.
• AWS revealed plans for a dedicated AI agent marketplace launching July 15, where Anthropic and other partners will sell their agents directly to enterprise customers.
• Former Intel CEO Pat Gelsinger created a new benchmark that measures whether AI models support human flourishing, adding “Faith and Spirituality” to standard well-being categories.
• Harmonic, the AI startup tackling math problems that stump current models, secured $100M from Kleiner Perkins with Robinhood CEO Vlad Tenev as executive chairman.
• METR released new research that conducted a randomized controlled trial with 16 experienced open-source developers and found that using AI tools like Claude 3.5 Sonnet actually took them 19% longer to complete real tasks on their own open-source repositories, yet they still believed they were 20% faster.

For the latest AI deep dives, check out our Explainer articles here!

FROM OUR PARTNERS

Vellum AI: Build, test, and monitor AI in production

Tired of prototypes that go nowhere? You’re not alone. That’s why Vellum’s team built a single place where engineers and non-engineers can build, evaluate, and ship AI workflows that work in production. Get free access here!

Intelligent Insights

• In a survey of 1K+ C-suite execs (CEO, CFO, etc), Icertis found 90% claim increased business pressure, with 46% struggling to show AI ROI and 46% racing to keep up with AI innovation. Executives cited AI washing (51%) and unclear AI use cases (44%) as investment decision barriers, while 90% expected tariffs to hurt profits.
• Researchers designed an AI that can “autonomously discover and exploit vulnerabilities” in smart contracts—put another way, it can steal crypto (paper).
• AI is increasingly making Apple look like a loser to the stock market, with the stock down 15% due to concerns over tariffs and its ability to compete with AI, and some analysts believe acquiring Perplexity is its only shot at keeping up.
• A basic security flaw in the McDonalds hiring AI called “McHire” (built by Paradox) exposed millions of applicant’s sensitive data to hackers. Why? Because of the password “123456”.
• Check out this fascinating argument that language models like ChatGPT are actually the perfect validation of 1980s French Theory—they prove language can generate meaning purely through statistical relationships between signs, without any understanding or consciousness, exactly as structuralists claimed before anyone imagined such machines could exist.

Check out all of our Intelligent Insights from June here!

A Cat's Commentary.