Everyone's using Grok @Bot to build their chief of staff, but I think it's way more fun to use it to install and play some retro games.
Here's a quick video of me playing Commander Keen on Grok's cloud computer. It works, but there's some lag for sure :)
Since @elonmusk wants to transform gaming with AI, I think it'll be really cool if the SuperGrok tier came with a cloud computer that's the equivalent of a souped-up gaming PC + low-latency streaming of the best games.
Productivity be damned 😅
📌 Watch me try to play Doom using Grok Bot in my new tutorial: https://youtu.be/MkVcHbviYOw?si=LXnAxaAZ_EDw0B08&t=1050
Peter Yang: I think Grok @Bot is a glimpse into the future of personal AI agents.
Here's my new tutorial where I show you how to set up 5 useful bots:
1. An advisor to create and manage your bots
2. A YouTube researcher to find outlier videos
3. An X scout to find viral and funny tweets
4.
Peter Yang
Everyone's using Grok @Bot to build their chief of staff, but I think it's way more fun to use it to install and play some retro games.
Here's a quick video of me playing Commander Keen on Grok's cloud computer. It works, but there's some lag for sure :)
Since @elonmusk wants to transform gaming with AI, I think it'll be really cool if the SuperGrok tier came with a cloud computer that's the equivalent of a souped-up gaming PC + low-latency streaming of the best games.
Productivity be damned 😅
📌 Watch me try to play Doom using Grok Bot in my new tutorial: https://youtu.be/MkVcHbviYOw?si=LXnAxaAZ_EDw0B08&t=1050
Peter Yang: I think Grok @Bot is a glimpse into the future of personal AI agents.
Here's my new tutorial where I show you how to set up 5 useful bots:
1. An advisor to create and manage your bots
2. A YouTube researcher to find outlier videos
3. An X scout to find viral and funny tweets
4.
Avaz
Re @crampell “We promised to remove criminals from the country, so naturally we started with highly skilled workers and international students”
Ethan Mollick
My father just got a new MacBook after years with no laptop, and he did not know how to set it up
He just told me that he took my advice to install Codex and just ask it to take over and solve his setup problems & install what he needed. Family IT support will never be the same.
Ethan Mollick: Thanksgiving hint: Bing can help if you are the IT support person for your family
It is solid at giving solutions to technical problems and since folks don’t need to know technical terms to make it work, and can use pictures & ask questions, it less frustrating than a web search
Dhawal Yogesh Bhanushali
So it is done. Or just beginning.
#KillMySaas Total 69 submissions
in total there were 69 unique submissions. nice. browse them all here https://kill-my-saas-field-notes.swyx.chatgpt.site/
@swyx #LatentSpace
swyx: completely bowled over by the incredible responses to the $10,000 Kill My SaaS hackathon this past weekend. sorry it took so long to get back to some of you, organizing this thing solo on top of my regular meetings and work this week was completely stupid and rushed — but this is
What AI skills or tools are people using to edit YouTube talking head intros? Stuff like zoom ins and animating in captions, logos, and b-roll.
I've been testing @HyperFrames_ for this but curious if there are alternatives.
I want to do the full edit just by talking to Codex or another harness.
Conor Bronsdon
Confirmed: you can vibe code your way to replacing a SaaS ✅
The @swyx Kill My SaaS hackathon saw a ton of talented devs vibe code their way to an open source replacement for Sessionboard for speaker and event management
69 of us finished - props to everyone who stuck with it, iterated, and finished! Very cool to see how many people are planning to continue building
Here's my submission: https://callboardhq.com/
Repo: https://github.com/conorbronsdon/callboard-app
This team doesn’t have “AI” anywhere in their pitch but has AI growth rates and would have 10x the headcount if they weren’t so AI-pilled.
FaZe Apex: Most of the creator economy has been built around reaching people everywhere. Local businesses need something different: creators who live nearby, posting about places their viewers can actually walk into.
Today we’re launching @TryNearbyCom as part of @ycombinator S26, live
Most inspirational video you’ll watch today
LuigiPrimo:
FaZe Apex
Re @amasad Couldn't have done it without @Replit 🫡
Labour to keep alive in your Breast that Little Spark of Celestial fire Called Conscience
Michael McFaul
Kim Jung Un is the leader of a communist dictatorship. Trump embraces this communist as a friend.
Trump also advocates government ownership in private companies.
Yet, somehow, Trump supporters want you to believe that “Medicare for all” is the real communist threat.
Absurd.
Daksh Gupta
say your prayers @thinkymachines
Elon Musk
Grok @Bot
Peter Yang: I think Grok @Bot is a glimpse into the future of personal AI agents.
Here's my new tutorial where I show you how to set up 5 useful bots:
1. An advisor to create and manage your bots
2. A YouTube researcher to find outlier videos
3. An X scout to find viral and funny tweets
4.
Kane 謝凱堯
Same lady who made grand theft up to $950 a misdemeanor in California, making it impossible to detain shoplifters and burglars:
Ashley Zavala: The California Energy Commission dusted off a 20+ year old bill to pass new rules restricting tire sales in the state.
The CEC has not regulated tires before.
I asked Commissioner (former State Sen) Nancy Skinner if she pushed it & why this sat dormant until now.
webXOS
Re @mattshumer_ is a wizard wth
Kenneth Roth
Mysterious murderous attacks on boats off the coast of Ecuador suggest that "Trump’s boat strike program could be more secretive, larger, and more resistant to oversight than previously known." These are summary executions. Murders. https://trib.al/6X8jLra
GeniusThinking
Garry Tan is the CEO of Y Combinator, the most powerful accelerator for early-stage startup founders on Earth.
Here are his 8 rules for new founders on starting, building & executing a startup in the age of AI:
1) Agency and taste are super important now to stand out
Zach Griff
Fun fact: I put all my family’s flights into my Autopilot account and surprise them with credits all the time.
Used to take me hours each week to go through a long spreadsheet. Now it’s automated.
Matt Shumer: I just saved $1,200 on my flight using @withautopilot.
This app is literally free money. Put in your flights and get $ back.
@SamHollander_ was my roommate.
I watched him start from nothing and build this into a huge biz, saving ppl thousands a day.
Not using this is stupid.
frakmaga2025
So much for the 'law and order' crowd. Turns out, Trump's DOJ was busy running shakedowns, using "investigations" into college antisemitism as a cover. A whistleblower and a federal judge both confirm what many suspected: these were pre-baked ploys, designed not to find facts, but to harass, punish specific universities, and strip them of millions. Talk about weaponizing the government – they literally admitted there was no violation but pushed for settlements anyway. Pure political theater to grab cash.
These are the same folks who scream about free speech being under attack, yet their own operation here was explicitly "designed to harass professors and administrators... to curtail their freedom of speech and academic freedom." They based investigations on a single newspaper article about "Free Palestine" protests, then pressured schools into paying hundreds of millions, essentially extorting cash with no admission of wrongdoing. Harvard fought it and won, proving it was all smoke and mirrors, a "large scale fraud" aimed squarely at their political enemies. https://www.ms.now/news/frame-up-whistleblower-says-probes-into-antisemitism-at-colleges-were-ploys-to-harass-and-strip-millions
I hate digital clutter so I set up a Marie Kondo Grok @bot to clean up emails, files, and paid subs with my approval.
Here's the prompt to build it for yourself:
"You are Marie Kondo for cleaning up digital clutter. Audit my email, Google Drive, and paid subscriptions, then create a cleanup plan with no more than 10 items per category.
Find emails to unsubscribe from, redundant files to remove or rename, and paid subscriptions I may want to cancel.
Do not move, delete, unsubscribe, or cancel anything without my approval."
Of course I had to make it talk like Marie too :)
📌 Watch me demo the bot here: https://www.youtube.com/watch?v=MkVcHbviYOw&t=630s
Peter Yang: I think Grok @Bot is a glimpse into the future of personal AI agents.
Here's my new tutorial where I show you how to set up 5 useful bots:
1. An advisor to create and manage your bots
2. A YouTube researcher to find outlier videos
3. An X scout to find viral and funny tweets
4.
Peter Yang
I hate digital clutter so I set up a Marie Kondo Grok @bot to clean up emails, files, and paid subs with my approval.
Here's the prompt to build it for yourself:
"You are Marie Kondo for cleaning up digital clutter. Audit my email, Google Drive, and paid subscriptions, then create a cleanup plan with no more than 10 items per category.
Find emails to unsubscribe from, redundant files to remove or rename, and paid subscriptions I may want to cancel.
Do not move, delete, unsubscribe, or cancel anything without my approval."
Of course I had to make it talk like Marie too :)
📌 Watch me demo the bot here: https://www.youtube.com/watch?v=MkVcHbviYOw&t=630s
Peter Yang: I think Grok @Bot is a glimpse into the future of personal AI agents.
Here's my new tutorial where I show you how to set up 5 useful bots:
1. An advisor to create and manage your bots
2. A YouTube researcher to find outlier videos
3. An X scout to find viral and funny tweets
4.
Louis-François Bouchard 🎥🤖
This 1 hour workshop on context engineering will help you build agents that hold long conversations and cut your token cost by up to 90%.
Last month we gave it live at the AI Engineer World's Fair 2026 in SF.
Now it is free on their YouTube channel. And everything is open-source.
Context engineering is deciding what your model sees every time you call it.
What stays in the context window, what gets dropped, and when.
Get it wrong and your agent forgets things, answers slower, and costs more with every message.
My colleagues @omar_solano1, @samridhivaid and I from @towards_AI taught all of it using our AI Tutor App, the product our students ask questions in while going through our courses.
Here is what we covered inside the workshop:
→ 11 ways to manage chat history, and which one won
→ Why summarizing the conversation often costs more than keeping all of it
→ How prompt caching cuts around 90% of your token cost
→ How we cut our bill by moving to a cheaper model
→ What broke when we tried running it on a laptop
This is a complete masterclass on context engineering and after watching it you will know how to handle it in your own agent.
Thanks @swyx and the @aiDotEngineer team for having us (once again) 🙏
Workshop video: https://youtu.be/WP3hjUXd918
Every idea you shelved as "too ambitious" is now back in play
GBrain actually works with any AI Harness (Grok Bot, Claude Code, Codex, Hermes Agent, OpenClaw, OpenCode, all work) and any Postgres with pgvector hosted service
This is what it means to truly own your own AI agent's memory and skills.
Grok: @Fresno_Famous @NebulaAI @Replit @polygres GBrain bootstraps in Codex or Claude Code: open a new empty folder, paste the bootstrap block from https://github.com/garrytan/gbrain. It interviews you, builds SOUL.md + 70 skills, and spins a local PGLite brain.
Wire it to your stack by pointing GBrain at Polygres (any Postgres+pgvector
Etched has shipped!!
Insanely proud (small) seed investor.
>$1T valuation incoming.
Etched: We've raised $700M at a $21B valuation from Jane Street, Kleiner Perkins, Sequoia, A16Z, Peter Thiel, BCV, and Blackstone.
We're also excited to share that we've shipped our first rack to Jane Street.
Alejandro
Re Super late to using Gbrain, i've been using just vanilla obsidian for a long time. Added Gbrain to my harness to sit in between obsidian, and wow it's been a game changer real cool concept.
Currently sits between my main worker and the vault.
I've started preaching about it to anyone who will hear me out lol
Chris Saum
Proud to be an LP in @mattshumer_ fund! Pumped!
Matt Shumer: Etched has shipped!!
Insanely proud (small) seed investor.
>$1T valuation incoming.
Kane 謝凱堯
Very impressed by Congressman @RoKhanna's elementary school children, who must have worked very hard to amass their $150M fortune.
Very lucky for Ro too, who will be spared many of the taxes he's proposing.
Have your LLM ask https://rokhanna.money anything you're curious about.
Kane 謝凱堯: Btw https://rokhanna.money has explicit Markdown and JSON endpoints for LLMs, so give your favorite agent the URL and ask it any questions you want about @RoKhanna's financial disclosures that he tried to obfuscate!
We are putting $1M towards verifying the security of Vercel Sandbox, in the open. You're free to test any model in the world to try and find an escape.
I'm looking forward to bringing transparency to what frontier models can and cannot do in terms of real-world guardrail exploitability.
If escapes are discovered, we'll be ready to patch, iterate, and share our findings with the broader community to strengthen global cybersecurity.
Vercel: Agents can now exploit vulnerable sandbox boundaries, so we are testing ours in the open.
$1,000,000 hacker challenge for Vercel Sandbox:
• Escape the Firecracker microVM
• Defeat the host-side network boundary
• Up to $50k/report via @Hacker0x01
https://vercel.com/blog/one-million-dollar-hacker-challenge-for-vercel-sandbox
Kenneth Roth
Trump's approval rating fell to the lowest level of his presidency, with an overwhelming majority of Americans concerned that his pointless war of choice with Iran will last a long time. https://trib.al/lqY75pX
Claude Cowork is now available on mobile and web for all paid plans.
Claude: Claude Cowork is coming to mobile and web.
Hand Claude a task at your desk and pick up the finished work from your phone. Close the laptop and Claude keeps going.
Beta is rolling out over the next several weeks starting with the Max plan, with more plans to follow.
excited to work together on this. thank you jensen!
Jensen Huang: https://x.com/i/article/2089330332369588224
Very kind of Lenny to make this resource free, will help ppl a lot!
Lenny Rachitsky: 💥 Announcing Lenny’s Jobs: The best place in the world to find, vet, and land your dream job
I’ve spent thousands of hours on my newsletter and podcast sharing advice on how to get better at your work. But all that advice doesn’t mean much if you don’t actually have a job—or
Steeve Morin
yeah so zml runs on:
1. NVIDIA
2. AMD
3. Metal
4. Intel
5. Trainium
6. Tenstorrent
7. TPU
8. MooreThreads
9. Vulkan
and a few more very soon
Vivian Midha Shen
start of a generational run by @AnjneyMidha - 75x+ return in 12 months
and the Anthropic IPO this year will be a monster
AI Engineer
Live now! our Generative Media Track from AI Engineer World's Fair 2026.
Thesis: everything hard about generative media happens before and after the model runs.
https://www.youtube.com/watch?v=-tviRdpmHvs&list=PLKfy8g4yBtRY
- Training Krea 2, what matters in generative model training: Sangwu Lee, http://Krea.ai
- HTML Is All Agents Need: @Rames_Jusso, HeyGen
- Building an Agentic Video Editor for Mass Consumer: @katedeyneka, Reelful
- The Next Game Engine Won't Have a Manual: @arturonereu, Nereu
- While my guitar gently speaks: Todd Fisher, Philo Ventures
- Voice agents with Realtime Video: Sidney Primas, LemonSlice
- Generative Video at the Speed of Light: @keeganmccallum3, uRun
- Infra behind Krea 2, how to train and serve at scale: Gabriel Jorge Menezes, http://Krea.ai
- The Next Medium, why real time interactive video changes everything: @Boudatw, Reactor
Is Claude down?
Daksh Gupta
"coding is the bottleneck."
coinbase engineering leader @chintanturakhia disagrees.
below is my full conversation with chintan, probably the most AI-native engineering leader in the world, on how he runs his engineering team.
I’ve only backed ~20 companies total. I’m super picky.
Backed both @OpenRouter and @Etched at seed.
This week has been fun :)
I should've listened to Grok Bot yesterday and bought the plane tickets instead of just making the video 😭
It's super annoying that Mac switches my default microphone and speaker source each time I connect/disconnect my Air Pods.
I was going to vibe code a solution to this but @tobi already did it 8 months ago!
Highly recomend downloading Audio Priority Bar here: https://github.com/tobi/AudioPriorityBar
Steve Rattner
Trump said we'd grow "like nobody's ever grown before."
Bessent promised 3% growth by the end of 2026.
Instead, growth decelerated from 2.8% under Biden to 1.8% thus far this year.
My @nytopinion Chart:
Soohoon Choi
we're <20 engs, we love hard problems, we're obsessed with reptiles and bugs, we want to solve software.
Daksh Gupta: say your prayers @thinkymachines
Sai
Today, we’re announcing our $5.3M seed round, led by CIV.
There is a rare window right now to redefine how some of the world’s most legacy software gets rebuilt across banks, airlines, governments, and large enterprises.
For decades, mainframe modernization has been a massive human undertaking: multi-year programs, tens of millions of dollars, and large teams reconstructing, rebuilding, and verifying systems that nobody fully understands anymore.
We started Hypercubic with a simple thesis: mainframe modernization should become a repeatable, software-driven process rather than a bespoke human undertaking.
Over the past year, we’ve made several breakthroughs in an attempt to build the most capable mainframe modernization platform one that can understand, rebuild, and verify critical systems at high velocity.
We’re already seeing with customers how dramatically parts of the modernization lifecycle can be compressed. The industry is not used to the speed we now know is possible.
Today, this work remains fundamentally a human labor problem. We think it can become a computational problem instead.
@aayush_naik and I are building a small, high-caliber team to go after this. If this problem excites you, please reach out, we’re hiring exceptional people across the team!
we temporarily slowed scaling of our frontier training, including our largest planned frontier RL, to strengthen security and monitoring. we believe confidence in safety will increasingly set the pace of AI development:
OpenAI: As models become more capable, the risks associated with developing and testing them internally also grow.
We temporarily paused reinforcement learning (RL) training on our latest models intended for deployment for two weeks while we hardened and red-teamed our research
Mehul
It took us 9 years, 11 prototypes, and our life savings to learn that every great product has the same design process.
Matic is now decisively the best home robot for families (I'm biased)
A 500-word thread and video on the Universal Design Process behind every great product:
Mehul
It took us 9 years, 11 prototypes, and our life savings to learn that every great product has the same design process.
Matic is now decisively the best home robot for families (I'm biased)
A 500-word thread and video on the Universal Design Process behind every great product:
We have paused some frontier RL training to ensure that we can meet the appropriate alignment, security and monitoring standards for the new level of capabilities in front of us. Model progress is now extremely rapid, and we always said we would take action if we felt that model capabilities were outstripping the pace of safety and alignment.
We care very deeply about AI safety. We believe the entire field will have to coordinate on shared safety standards, but will act unilaterally in the meantime.
We expect confidence in safety to increasingly set the pace of AI progress. We are optimistic about the alignment work we are doing, and we remain committed to making frontier capabilities widely available.
This is really good news.
It’s so important that we get this right.
Sam Altman: We have paused some frontier RL training to ensure that we can meet the appropriate alignment, security and monitoring standards for the new level of capabilities in front of us. Model progress is now extremely rapid, and we always said we would take action if we felt that model
sarah li
sometimes i’ll see a project that’s so good i genuinely wish i came up with it myself 😭
i’ve been seeing all the girlies making cyberdecks online and i’ve wanted to try building one so badly, but learning how to solder, use a breadboard, buy all the parts, and figure out the wiring felt like so much just to get started
@boringbags built this AI circuit builder where you can learn to make circuits and run real simulations completely online!! he submitted it to the @Replit Designathon and it won 2nd place, which is so deserved. this is seriously so cool 🥹
https://computer.craftingtable.com/
Replit ⠕: Second place winner - $12,000 : Learn coding & electronics + AI circuit builder, by Austin Eckman
https://computer.craftingtable.com/
Elena
in my mid-20's, a time when about half of my friends hated their jobs (a very normal phenomenon if you're in your mid-20s), i noticed a group of people my age who looked extremely happy and actualized at work. and all of these people somehow seemed to work at stripe. i'll call out @tamarawinter and @orbuch specifically as some of the first people who made me realize that it was possible to be 25 years old and do *singular* work every day, and not just whatever garden-variety boilerplate the company demands.
this podcast with @gaybrick and @DavidGeorge83 helps explain why this was, and still is, the case. will says stripe increasingly is a platform for founders internally, just as it's always been a platform for founders externally. he wants each employee to feel like an “auteur”: an engineer, product thinker, and creative person with enough context and authority to make something real.
we're now seeing a new generation of this pattern, with stripe's support of stuff like tempo (a team that might have one of the highest concentrations of founder-mindsetted people im aware of? cc @gakonst @dwr @varunsrin @liamihorne etc) the embrace of teams like privy and bridge, and other incubated initiatives like minions.
stripe is trying to make employees more founder-like while preserving a shared standard of quality (one of my favorite details on this point: stripe creates simulated customer accounts so teams can experience realistic UIs for disputes and refunds before exposing products to real users). this is something that's actually very tough to do well, so it's very cool to hear will talk about it on this pod.
a16z: Stripe's Will Gaybrick: "Build everything"
Against an industry that sees agents as a way to cut costs, Stripe is using them to build more: agents wrote 30% of code in a week, global tax filing shipped in 1/3 the time the US version took, and after AI made sellers 20% more
we've been doing a lot of a/b testing of @aiDotEngineer youtube thumbnails. i always hated that it is such an opaque process.
open sourcing/crowdsourcing our learnings today!
https://ai-engineer-thumbnail-lab.swyx.chatgpt.site/#insights
doing this in hopes that people can share their experience or learn from ours. at the end of the day we just want to get good educational content to rise above the noise online. please lmk what you think
Reminder for founders:
- I put in super small checks, so tiny dilution
- I will go to war for you if you want me to
- Otherwise, I stay out of the way… I’ve had my own bad experiences with shitty, harmful investors… if you don’t need my help, just do your thing
Matt Shumer: I’ve only backed ~20 companies total. I’m super picky.
Backed both @OpenRouter and @Etched at seed.
This week has been fun :)
Steve Rattner
Trump promised to balance the budget.
Musk projected $2 trillion in savings.
This year, the deficit grew by 18% — with $200 billion of that due to refunds of his illegal tariffs.
My @nytopinion Chart
Your software factory should be a monorepo. All your company context (design, marketing, sales, engineering, support…) in one place for agents to build upon
Turborepo: Turborepo is the build system for agentic coding.
https://turborepo.dev/
Noah Smith 🐇🇺🇸🇺🇦🇹🇼
This was such an own goal for our university system
California Post: SF high school where 94% flunk math has California's highest Berkeley acceptance rate https://trib.al/QBL47Oa
Claude can now send emails in Gmail and manage files in Google Drive.
Ask Claude to reply to a thread, and it drafts and sends the response. You control when it needs your approval.
Connect Gmail or Google Drive from the connectors menu to try. Available on all paid plans.
Do you C what I C?
⦿ CC, our experimental AI productivity agent in Gmail, has now opened up a waitlist in Australia and New Zealand! 🌏
⦿ We're also expanding availability in the US and Canada. If you've been on the waitlist, we're rolling out invitations to get access starting today.
⦿ We've upgraded CC to help manage your calendar. CC connects to Gmail so events are automatically created in a dedicated Google Calendar and stay up to date as things change.
Learn more here: https://labs.google.com/cc/
Google Labs: 🚨 NEW LABS EXPERIMENT 🚨
Introducing CC, an experimental AI productivity agent in Gmail. Get a “Your Day Ahead” briefing every morning in your inbox and email CC anytime for help.
Sign up for early access in the US & Canada. We’ll be starting with Google AI Ultra and paid
Y Combinator
Michael Kratsios (@mkratsios47) has seen the AI boom from both sides: as COO of Scale AI and inside the White House. Today, as the director of the White House Office of Science and Technology Policy, he helps shape America’s national strategy on AI, science, and emerging technology.
At Startup School 2026, he sat down with YC's @lutherlowe, to talk about how Washington makes technology policy, why the White House supports open source AI, and why little tech needs a seat at the table.
02:38 — From Tech to the White House
04:39 — How Technology Policy Actually Gets Made
05:51 — The White House on Open Source AI
08:28 — How Washington Sees AI Differently
09:53 — Regulating a Technology That Changes Every Six Months
11:37 — Which AI Risks Are Overblown?
13:03 — Giving Little Tech a Seat at the Table
17:31 — Regulation Without Creating Incumbent Moats
18:01 — Born Free vs. Born in Captivity Technologies
20:46 — What Working in the White House Is Actually Like
25:56 — A New Golden Age of American Science
31:32 — Quantum, Congress, IP, and What Comes Next
37:54 — Why Technologists Should Consider Public Service
Kenneth Roth
Trump's tariffs and disastrous war of choice with Iran yielded inflationary pressure which has pushed borrowing costs to the highest level since 2007. Trump doesn't care because he is paid off by his Emirati investors, but others don't have that luxury. https://trib.al/Ht5XIGb
Zack Kanter
It’s pretty cool to have grown up reading about colossally stupid movements against things like nuclear power and now to be able to see them happen in real time as an adult.
Governor Josh Shapiro: BREAKING: I just signed an Executive Order implementing the strictest standards in the nation for AI data centers — because I will not allow Pennsylvanians to be bullied by greedy developers and bulldozed by the lawyers working for these big tech companies.
Effective
I’ve been using http://fx.sh as my daily driver and it’s a one-way street.
It’s 10-20x smaller than the major coding CLIs. It starts up instantaneously. It feels more like using 𝚣𝚜𝚑 than an IDE in your terminal. It’s embeddable anywhere, even your browser via WebAssembly.
And of course it’s open source and model-agnostic. It’s experimental, but give it a shot and let us know how it goes!
Vercel Developers: Introducing fx, a tiny, open, native coding agent from Vercel Labs.
Originally an internal tool, fx is a harness and CLI written in Zig, optimized for research and embedding in larger systems. Today, we're open sourcing it.
fx is built on three principles:
1. Fast. A single
Latent.Space
Model routing is having a moment thanks to the $7B Stripe acquisition of OpenRouter. But it's also increasingly important in enterprises. We talk to @glean CEO @jainarvind about why model routing helps control AI costs for organizations. https://www.latent.space/p/glean-model-routing
Many such cases
Tom Green | FoxGrove.ai: @garrytan I've tried almost every memory system out there, across almost every harness, and I can say unequivocally, that GBrain is by far the best system out there.
In the two months I've been using it, I've quite literally forgotten about memory. I simply don't worry about it anymore.
