An Engineer's Log

We just launched Hosted OpenClaw Instances with full managed hosting, dashboard, terminal access etc. all for $0.99/mo And it just takes 30 seconds to set up! agent37.com/openclaw

@cailynyongyong @minmax0501 @getyourmomo Very cool! Thanks for sharing

HOW TO GO VIRAL ON X: complete basics easy mode with first guest as @minmax0501 00:00 Intro 00:50 @getyourmomo 01:16 First guest intro 02:16 Warm up account 02:51 Tip 1: tag communities 04:19 Tip 2: tag famous accounts 05:06 Understanding X audience 06:00 Things that didn’t work 06:28 Linkedin vs X 07:14 Tip 3: post everyday 08:20 Do videos work? 09:03 Tip 4: post successes 09:44 Do multiple channels work? What’s the difference? 10:40 Tip 5: wait for random viral posts 12:00 What’s your brand message? 13:10 Tip 6: Copy viral templates from competitors 13:25 Tip 7: Follow the trend 14:35 How to structure demo videos 15:10 1 second hook 16:13 First 5 seconds matter 17:48 1 feature / video 19:17 Story telling 21:32 Writing good script 22:47 Yeet

@Teknium Wow nice!! 😍 The UI dashboard was a much needed thing

So much in this release but the one many have been waiting for above the rest, the GUI dashboard! Manage and monitor your Hermes Agent with a GUI Local Web Dashboard with `hermes dashboard` command to start it!

@tech_savvyact Thank you, the goal is to make this affordable to masses :)

@an_engineer_log $3.99 is aggressive pricing. The 'just working' feeling is rare. Most agents are 80% there and break on the last 20%.

I’ve spent the last two weeks testing Hermes against OpenClaw. With OpenClaw, there are just too many ways for things to break. Hermes is the first agent I’ve used where you actually get the feeling of it "just working" You can now spin up Hermes on Agent37 for just $3.99. • 1-click integrations for 1000+ apps (Gmail, Calendar, Notion, etc.) • Pre-configured LLM (swap any OpenRouter model) • Full GUI access to your files and live browser Demo 👇

@michael_chomsky @AnthropicAI Interesting.. I kinda brushed it off.. but what specifically about this do you find great ?

Anthropic managed agents is a work of art. It's solving a lot of problems nobody has even come close to solving. It's not just one or two innovations, it's like 20. Some serious thought went into this. Incredible work by @AnthropicAI. If you're building in the space, I'm implore you to take notes.

@paulg @Ric_RTP so only sometimes?

@Ric_RTP No, I did not tell colleagues Sam lied to us all the time.

The journalist who took down Harvey Weinstein just spent 18 months investigating Sam Altman. And what he found out is genuinely insane: The people who built OpenAI went on record saying he can't be trusted with the future of humanity. A Microsoft executive even compared him to Bernie Madoff. This isn't just some hit piece. It's 100+ interviews, secret memos, HR documents, Slack messages, and private notes that had never been seen before. Here's everything you have to know about Ronan Farrow's investigation: Ilya Sutskever, OpenAI's former chief scientist and CO-FOUNDER, compiled 70 pages of internal evidence against Altman. Slack messages. HR files. Behavioral analysis. The word at the top of his list of Altman's "consistent patterns": lying. He sent the documents as disappearing messages because he was "terrified" someone would find them. They became legendary in Silicon Valley. Insiders just call them "the Ilya Memos." Dario Amodei, another co-founder who left to start Anthropic, kept his own private notes. One line: "The problem with OpenAI is Sam himself." Paul Graham, the man who RECRUITED Altman to run Y Combinator, told colleagues Altman had been "lying to us all the time." Multiple YC partners had complained about Altman's behavior by 2018. He was effectively forced out in 2019 despite publicly claiming for YEARS that he left voluntarily. Former board members described him as "unconstrained by truth." And the investigation found that Altman reportedly lied to the board about obtaining safety approvals for some of ChatGPT's most controversial features. That's the man running an $852 billion company with 900 million weekly users and a Pentagon contract. But here's where this gets really crazy: The New Yorker investigation dropped on Sunday. SAME DAY, Altman publishes a 13 page policy paper proposing robot taxes, a public wealth fund, and a four-day workweek. The most ambitious social policy document in OpenAI's history. Dropped within HOURS of the most damaging article ever written about him. That's not coincidence. Monday: Elon Musk files a court motion demanding Altman be REMOVED as CEO. He wants the for-profit conversion completely unwound. Then Friday at 3:45 AM: a 20yo throws a Molotov cocktail at Altman's San Francisco mansion. It bounces off the house. Lights the gate on fire. An hour later, same guy shows up at OpenAI HQ threatening to burn the building down. Police arrest him on the spot. Nobody was hurt. But within hours, Altman posts a photo of his husband and 1yo child on his blog. Writes that he hopes the image "might dissuade the next person." Then blames the New Yorker article for making things "more dangerous" for him. In 5 days, Altman went from the target of the most devastating investigation in tech history to the sympathetic father whose family was attacked. Now anyone who criticizes him has to do it in the shadow of a firebombing. The New Yorker spent 18 months building the case that Altman is dangerous. Altman turned it into the reason HE'S in danger. And none of this changes what Farrow actually found: - The co-founders don't trust him - The former board doesn't trust him - The chief scientist documented 70 pages of evidence and was too scared to send them through normal channels - Paul Graham says he was lied to - A Microsoft executive put him in the same sentence as Madoff The trial starts in 16 days. If Musk wins, the for-profit conversion gets unwound and Altman is removed. If Altman wins, the man that every person who helped build OpenAI has publicly warned about gets permanent, unchecked control of the most powerful AI company on Earth. Either way, one thing is now undeniable... The people closest to Sam Altman are the ones screaming the loudest warnings. And this week proved he knows exactly how to make sure nobody listens. Peak manipulation.

@shiri_shh Hahaha, eventually we're all gonna be stepping on each others foot 😂

Lovable CEO realizing their biggest model provider is now their biggest competitor.

So I just tried this and a few things stood out as blockers. Is there a way to skip onboarding entirely on first load ? since it's pre-configured in our environment . Second, I don't see a way to use gpt-5.4 for example via codex subscription which I use. Seems to force to use API key.

@an_engineer_log @Teknium Yes, supports many of the core slash commands already. My plan is for it to support all the rest of the slash commands that the CLI does soon.

@DantesClown Yes, it's very simple to migrate. It has a GUI file browser and you simply download your entire .hermes folder and have it locally instead :)

Nice looking product--I am running OpCl and Hermes on a macmini (base with 16gb RAM)...I am curious about this hosted option (for my wife who is curious but would want her own agent) but wonder if in the future she decided to migrate to a beefier machine, am I able to move my hosted agent and all its files and memories to a private home server? (its low prob since she is not technical but I just wonder about the future path options)--thanks

@dragosroua Very cool mate!

Or 5 distinct personas of AIGernon for $9.99

@moulougueta @NousResearch Very nice!

Deploy @NousResearch Hermes Agent in a minute using Clawnetes No terminal Chat with your Hermes agent using Clawnetes chat interface Chage the models with just a click Download now from aimodelscompass.gumroad.com/l/clawnetes

@nesquena @Teknium Does it also have the slash command support already btw ?

@an_engineer_log @Teknium Excellent, excited to have my web UI available to people that are getting setup with Agent. The web UI now supports the entire Hermes onboarding flow in your browser too!

@nesquena @Teknium Oh nice, I'm going to try it out tomorrow !

@tibo_maker Thanks Tibo ❤️ I'm a big fan of your products!

@an_engineer_log looks cool 🙌

@nesquena @Teknium Thank you, this looks great !

@an_engineer_log @Teknium Added the new hosting section! nesquena.github.io/hermes-webui/i… - nice addition!

@nesquena @Teknium Btw, it would be great if you could add a Hosting section here and add include agent37. I think that would be valuable for people looking for hosted options 🙏

@an_engineer_log @Teknium Great writeup! For those newer to Hermes and what it can do - or curious about Hermes vs OpenClaw vs Codex? Get a clear look where things stand and head to head comparisons: nesquena.github.io/hermes-webui/

@garrytan Hey @garrytan Here's another easy way to try OpenClaw and Hermes side by side :) x.com/an_engineer_lo…

If you haven't tried Claw/Hermes I just gave you the easiest, best, guided way to get my setup going for yourself. Here it is, it's open source, and would love for you to try it. github.com/garrytan/gbrain

Just launched GBrain v0.8.0 If you have it installed, you can just ask your Claw/Hermes to upgrade to the latest GBrain and we'll automatically ask if you want to install your Voice WebRTC endpoint and Twilio number It's a true mega brain-trip to talk to your agent directly.

@pepethehandsome It's been on our backlog for a while. I'll add this in a day or two :)

@an_engineer_log Really cool, will definitely try this. Would be cool if there was a backup option to put old agents on the shelf until you need them again, so they are not running pointlessly in the background

@bridgetdef ❤️

👍♨️🌡️🥵👍

@nesquena @Teknium Oh wow! I didn't realize. I'll test it out. Mind if we host it on Agent37 for our users if it works well ?

@an_engineer_log @Teknium Here’s what it looks like (has many themes)

I've been doing my workflows on both OpenClaw and Hermes is really great as a harness and less prone to errors, a lot more stable and breaks less. When using openclaw you kinda have to brace yourself for the bugs. Of course OpenClaw has a lot more app integration support so there is that trade off.

@nesquena @an_engineer_log @Teknium i'm still trying to wrap my head around hermes and openclaw, somebody explain it to me like i'm five, we use snowy ai at work and it's just a black box to me