Head of editorial at https://xitter.cf/t.co/UNIn3ajRjC / @latentspacepod Founded ReadWriteWeb (2003–2012) Internet historian at https://xitter.cf/t.co/ZlyypsMmpr 🥝 in 🇬🇧
World Wide Web
Joined January 2010
- Tweets14.3K
- Following2.5K
- Followers8.8K
- Likes9.5K
Pinned Tweet
Today's my first day in a new role: Head of Editorial at Latent Space (@latentspacepod).
From ReadWriteWeb to The New Stack, I've always loved being at the cutting edge of tech x media. So I'm thrilled to join @swyx & his team. We have big plans! latent.space/
Funnily enough, we're making the opposite move: from Substack to a more flexible, creative and web-friendly platform in Beehiiv, where we can Make Our Website Great Again.
I don’t think there’s any great mystery here. They’re outsourcing discovery and monetization to Substack because there’s not enough potential value to do it in-house. defector.com/nobody-wants-to…
One of the frustrating things about the Newsletter Era is a real lack of creativity in visual design these days (Subslop?). Check out my Mastodon/Bluesky alt account Classic Web if you want a taste of what the web once was: indieweb.social/@classicweb
Richard MacManus retweeted
There's not enough buzz around Claude Mods.
Read this Claude Mods article if you want to:
- Understand Claude Mods basics
- Install and use the most useful mods today
- Ponder where the agent-human interface will go
For @latentspacepod.
I totally missed that Foursquare was acquired! It was one of the most fun apps of the Web 2.0 era. My personal memories of using it: cybercultural.com/p/foursqua…
Foursquare was acquired last month. The outcome was disappointing. It's taken me a while to write this, but I think I'm finally done grieving it. Here are some thoughts on our final years, an effort to honor the team who helped build & maintain FSQ's legacy, and my attempt to finally get some closure. teendrama.substack.com/p/som…
Richard MacManus retweeted
First AIE Code Summit speakers:
@poteto, Principal Engineer @ SpaceX
@steve_yegge, Engineer, Writer, Advisor
@zachbruggeman, Principal Software Engineer @ Ramp
Nov 10-12, SF. Come hang out with us. Apply at ai.engineer/code
ALT AI Engineer Code Summit speaker announcement featuring Lauren Tan, Principal Engineer at SpaceX; Steve Yegge, Engineer, Writer, Advisor, Independent; and Zach Bruggeman, Principal Software Engineer at Ramp. Black and white artwork with three speaker portraits and yellow text: November 10–12, SF. Apply at ai.engineer/code.
Richard MacManus retweeted
Linus Torvalds to the anti-AI crowd: stop using compilers if you want to prove they’re REAL programmers. LOL.
“you had better not use an editor to write that commit message either.
The kind of expectations *I* have is that people do their best and I expect people to understand what they are doing.
But that understanding has very little to do with what tools they use.
And quite honestly, in my experience, the people I respect and call intelligent are not the ones that shy away from using new and interesting tools.”
It's been 3 days packed with learnings at @aiDotEngineer NYC. Here's my analysis of one of the biggest trends, now that frontier lab models are starting to 'cross the threshold' into work other than coding. I cover OpenAI, Anthropic, & 2 finance firms. latent.space/p/crossing-the-…
At AI Engineer New York, Anthropic's @cjav_dev & @HarrisonStall from @ClaudeDevs drew on lessons from Unix and Linux to explain how agents can tackle bigger projects without overwhelming developers. latent.space/p/claude-dynami…
Richard MacManus retweeted
Here's the docs for dynamic workflows for Claude Managed Agents: platform.claude.com/docs/en/…
Richard MacManus retweeted
Flows are pretty magical; couple lines of config in an API call and you unlock the power of complex, Claude orchestrated agent topologies for your hardest tasks.
Give it a try today and let us know what you think!
Richard MacManus retweeted
world’s first full length workshop on claude flows
@cjav_dev and Harrison Stall are kicking off @aidotengineer bright and early on a monday morning to a PACKED room, standing room only #AIEngineerNY
@ricmac is on the case reporting for @latentspacepod and we’ll also publish asap. proud to work with @anthropicai to sponsor AIE NYC for the second year running!
Made it to NYC for #AIEngineerNY, but this weather is worse than anything going on in the UK right now.
Fascinating interview, great stories of TechCrunch days. And man, the days when blogging was cool seem like ancient history now! Also I'm wondering if I too was a 'paranoid', like Mike, as I don't think I was a psychopath or a narcissist :)
All the things I used to write about at The New Stack are now owned by Cloudflare. They're like the Google of the web dev market.
We are excited to announce that Deno is joining Cloudflare!
deno.com/blog/cloudflare
This guy…knows how to social media. The first Gen Z to usurp his millennial boss? As a Gen Xer, I’m rooting for it.
I asked Alexandr Wang where he falls in the AI safety debate now that he's one of the people responsible.
His answer surprised me: leaders of AI are all brilliant, but personal feuds are shaping how this technology gets built.
"We have to all put our differences aside, put our histories aside."
Full HUGE* Conversation with @alexandr_wang out now
I do prefer Anthropic's literary naming over OpenAI's astronomical naming. It also makes more sense as a progression in models: haiku (small poem) -> fable (large story). Both companies should make it easier to determine which model (and strength) to use for each task tho.
Richard MacManus retweeted
NEW: Top SaaS Vendors on Ramp (Oct 2026)
We clearly see AI spend pulled in two different directions.
On the downside, new and cheaper models are pushing AI spend down. Jev captured an entire percentage point of market adoption within a month (unheard of growth). And routers continue to gain.
On the upside, the increased adoption of expensive image and video generation models will push AI spend up, like Higgsfield and Runway.
Richard MacManus retweeted
Putting down some thoughts on what seems to be happening to the open-source harnesses, guys.
1. I found it weird that Hermes published an index showing the best models to use in their harness, and told their users, who according to OpenRouter largely use cheaper models, that they are doing it wrong. Barely any of these models make it to their index.
2. But then the index has Terminal-Bench Science and Terminal-Bench. I found this strange, because a general/personal agents harness is evaling itself on science and code benchmarks.
3. My back and forth with one of the cofounders made me realize where their heads are at. He has since written a post to break up with the personal agents positioning they fought hard to win from OpenClaw.
4. This sums it up: "Dumb Normies". OpenClaw launched the first agent harness. It's long forgotten because of how un-normie friendly it was. Hermes took over, but is largely being abandoned just like OpenClaw, because easier products with better UX showed up, i.e. Grok Bot, Instinct, Muse, Dots, etc.
5. It seems the open-source harnesses like OpenClaw and Hermes feel they have lost the consumer opportunity, or that they have no chance of winning it.
6. OpenClaw launched an Enterprise beta, clearly positioning itself as a technical agent layer for enterprise. I called this 4/5 months ago when I was in the community team, Hermes was beating our ass and Peter was making even weirder moves. I said to the team then that OpenClaw feels it has lost, so it's positioning for enterprise.
7. Now it's Hermes' turn to position away from general agent to specialized harness, in their case coding and science. Sad to see them give up on consumer, but I guess they have to make money somehow, so this makes sense.
8. But it's not obvious even that they will win those. @clairevo is a power user and she's since dumped OpenClaw/Hermes.
9. I still think open-source consumer/personal/general agent harnesses need to exist. I'm still a big fan of OpenClaw/Hermes, but I can see Grok Bot, Muse and Dot pulling me away, largely because of the UX.
10. @kunchenguid wrote some days back that we still need an open general agent harness. Seems we need new OSS companies, since the 2 flag bearers have given up. (xitter.cf/kunchenguid/status/210…)
Both OpenAI and Anthropic have been trying to shift their agent harnesses more into the cloud — with varying levels of success.
Well, who better to make coding agents fully cloud-based than two #Kubernetes creators, @cmcluck and @jbeda. They spoke with me about their new company, Stacklok. Check it out on @latentspacepod: latent.space/p/stacklok
