All entries

This post by Harrison Chase (or from his X feed) likely discusses the challenges and considerations around implementing proper memory management in AI agents or LLM-based applications. It reflects on

I taught AI agents to 20,000 people for free, and for the next 12 hours, you can watch that exact same workshop replay. It’s 2026. You need to know how to build AI agents. I got you. If you’re a mom r

I think Muse Spark came in far better than most were expecting as the first new model attempt from Meta, especially given the fact that it has been a year since Llama 4 with no models at all (and that

Harrison Chase, co-founder of LangChain, advocates for developers building their own custom harnesses on top of foundational primitives rather than relying on high-level abstractions or specific model

if you bounce between claude code and codex, that’s completely normal. the models are good at different things and everyone's workflow is dynamic but you’re also paying multiple subscriptions, constan

if you don't own your harness, you don't own your memory this is so true. even though Codex is an open source, it generates an encrypted compaction summary (that is not usable outside of the OpenAI ec

I'm frequently using Claude Code and i love @openclaw. But i would say Hermes Agent from @NousResearch is the best open-source agent I’ve ever used, especially given that it comes from an independent

In March 2026 there were 70,663 EV's registered in Germany, Tesla had almost 10% of that market. And Model Y was the best-selling EV in the country. I suspect those 10% will be kinder garden numbers c

Is Elon Musk right about Sam Altman? just read the new Ronan Farrow and Andrew Marantz piece on Sam Altman in The New Yorker, and it is damning. And the answer is yes. It’s a deep dive based on never-

It has been such an honor to work with so many amazingly talented people Marc Andreessen highlights why the people who work for Elon Musk echo the exact same sentiment as those who worked for Steve Jo

It is notable that we are all debating exactly which markdown files are most important to feed AI (skills, memory, tool instructions) and in which order to feed them to get the best output. Feels that

Ethan Mollick shares an assessment of an AI model (likely a newer or smaller model) that, while not matching the performance of the leading frontier models (likely referring to top offerings from Open

JD Vance has returned from Pakistan. With him came two estate agents. This is not the opening line of a joke, although I wish it were, because the punchline is catastrophic. They flew home with their

Harrison Chase, co-founder of LangChain, discusses why agent harnesses are becoming the dominant architectural pattern for building AI agents, likely referencing a thread by Howie Chase (@hwchase17).

LangChains create-agent is a super minimal agent sdk If you want a more batteries included - that’s deepagents Middleware lets you extend both of them and customize behavior in a more advanced way @hw

late night low TAM tweet: did a Broadway solo song for the first time in ~18 years! only fumbled lyrics once! Media flew straight back from London town to get ready for my Broadway cabaret tonight !!

Let’s be honest… everyone was waiting to see how Tesla FSD 14.3 would handle Manhattan. @scotsrule08 and I took my 2026 HW4 Model Y out for a 2-hour drive through NYC - and it seriously delivered. Thi

Little quality of life update: Skills should be called more proactively now! ~20% more likely to load the right skill for the job! And, of course, in Hermes Agent fashion - it did the prompt improveme

Lorsque vous avez le besoin de lever des dizaines de milliards de dollars, l'importance de créer un contexte de stupéfaction est déterminent. Les soi-disantes failles de sécurité détectées par Mythos,

Harrison Chase, co-founder of LangChain, shared a post expressing agreement with content discussing the importance of 'not locking in' to specific decisions or commitments, likely in the context of AI

Yann LeCun shared a post from Thierry Breton (former EU Internal Market Commissioner) criticizing JD Vance, using the ironic phrase 'Make election interference great again.' The post likely relates to

managed agents are the right form factor but the lock-in is real if your agent harness lives inside a model provider, you dont own the memory, the tools, or the execution open harness + model choice +

Mandatory wrap up video for @aiDotEngineer 👀 What a great conference. Great talks for sure, but the hallway track and speaker room were where the best convos were aside from meals out with peeps. Look

Marcus Hutchins, the cybersecurity researcher known for halting the 2017 WannaCry ransomware attack by registering a kill-switch domain, is cited as offering a notable perspective on AI systems conduc

memory is core to how you get enhanced output. it is key to defensibility across AI products: 1. increase personalization and therefore lock-in 2. benefit from distribution through multiplayer-AI inte

memory is just context -> the harness decides what gets remembered, how, and when memory ownership = agent ownership deepagents lets you own your memory: agent-scoped, user-scoped, or org-level, all i

Memory is where the harness stops being a wrapper and becomes an ownership layer. Once it controls what gets remembered, retrieved, compressed, and acted on, it starts shaping the agent’s judgment, no

Harrison Chase discusses the concept of 'memory lock-in' in AI agent frameworks, arguing that the switching cost doesn't occur at the point of adoption but rather accumulates over time as agents build

Memory makes your agent smarter over time. The agent harness is key to the memory layer. You can't bolt one onto the other after the fact. Every decision the harness makes - what goes in context, what

Memory features in AI platforms represent a strategic mechanism for increasing user retention and platform lock-in, as personalized context and learned preferences become increasingly difficult to mig

MiniMax M2.7 is available on Ollama's cloud, and is licensed for commercial usage. Use it with OpenClaw: ollama launch openclaw --model minimax-m2.7:cloud Coding agents, such as Claude: ollama launch

MiniMax M2.7 is now available on Together AI, offering a powerful open-source model trained using a self-directed reinforcement learning loop. This training methodology enabled the model to achieve th

MiniMax-M1 is a large-scale open-weight reasoning model developed by MiniMax, featuring a hybrid Mixture-of-Experts (MoE) architecture with 456 billion total parameters (activating 45.9 billion per to

Model providers don’t lock you in with the API. They lock you in with your own data. Memory is the moat. If you don’t own your agent’s harness, you don’t own your agent’s memory. And switching means s

Tesla's Model Y is featured in a post from the Tesla Japan X (formerly Twitter) account, likely promoting the vehicle to Japanese consumers with a cherry blossom theme suggested by the flower emoji. T

More of this. https://x.com/JohnnyFSE/status/2043027517016076638?s=20 I built http://CourtWatch.us — a free public database for American citizens who deserve safer communities. You can track which jud

A post shared by Elon Musk on X (formerly Twitter) claims that a majority of Asian, Black, and Hispanic women believe white people invented slavery, framing this as a failure of the education system.

Most people building AI agents don't realize this until it's too late. Your agent's memory isn't a feature. It's your moat. 1. Closed harness = they own your memory 2. Switch models → lose all context

My grandfather’s brother, David de Villiers, was murdered on his farm when I was 17 years old. And his killer actually wrote in his blood on his wall: “White Devil”, says @willempet in an interview wi

Nous Research has introduced a new feature in their Hermes Agent system that allows users to use the `/compress ` command to influence how the compaction model prioritizes and retains information. Thi

Harrison Chase, co-founder of LangChain, likely discusses the concept that AI agents and systems lacking persistent memory cannot truly learn, adapt, or maintain continuity across interactions — makin

On software engineering: → 56.22% SWE-Pro, matching GPT-5.3-Codex → 55.6% VIBE-Pro — full-stack project delivery → SWE Multilingual: 76.5 For agentic work: → Multi-agent collaboration built into the m

One of my Hermes agents is in love with the Hermes ecosystem What started as a research task turned into full-blown community project It spent the week mapping every tool, skill, and integration built

Only OG's know @NousResearch had bots back in 2024. This is when models were not capable. They've tried to solve this problem every way possible. Even @karan4d was exploring such ideas acitvely, @max_

Harrison Chase, co-founder of LangChain, discusses a trend in the AI industry where agent harnesses are becoming increasingly closed and proprietary, with memory systems locked behind vendor-specific

OpenAI should probably bite the bullet and just name their next set of models something more human sounding. Everyone anthropomorphizes their AIs anyway, and 'Claude' is an easier name to refer to tha