Kova

@nickbaumann_ Nice, I was just thinking about setting up something similar! Mac Mini as a small always-on infra node for reliable work and my MacBook Pro for flexible high-power execution.

My laptop has become a “satellite device” since I started using Codex from my phone. And my Mac mini has become the “home.” It’s clunky, but the end state feels more like how we’re going to be working in the near future: I’m currently running the Codex app on 2 devices: 1. my MacBook 2. my Mac mini My laptop isn’t reliably connected to Wi-Fi enough, so I keep a Mac mini on my desk that is always connected. When I kick off new threads from my phone, I start them on the Mac mini. When I’m working from my desk, I run them there too. The cool part is that I’ve added my MacBook and Mac mini as connected devices to each other. That means I can start and resume threads from either device. So if I’m in a meeting but want to continue a thread on my laptop that was started on my Mac mini, I can do that. I’ve also set up mutual SSH for Mac mini <> MacBook, so files are easy to access from either side. It’s not fully seamless yet, but the model works. What this means: - I have an always-on Codex that is accessible from my phone, with its own dev environment - All threads are always accessible from any of the 3 devices - I can run heartbeat threads that stay on 24/7 It’s a little makeshift today, but the shape of it feels very real to me: Codex is no longer tied to whichever computer happens to be open in front of me. It starts to feel like something I can stay connected to across whatever device I’m using.

@BuildwithPublic @public Any plans on adding financials/fundamentals for stocks? I’d be willing to pay extra for that to have all my market data from one source 🙏🏼

@ApollosMission Overwatch is hot rn

Call of Duty is dead. Arc Raiders is dead. Fortnite is dead. Battlefield is dead. GTA is still not out. What is anyone even playing anymore these days?

Definitely an ugly quarter for $COIN. Still very exposed to crypto cyclicality, and the numbers showed it. But I don’t think this breaks the long-term bull thesis: stablecoins, on-chain financial infra, and eventually agentic commerce. For my book: holding, not adding yet. Need to see more of the thesis show up in the revenue mix.

@BuildwithPublic @public You guys are killing it 💪🏼

We just shipped a CLI for the @Public API. Traders who live in the terminal — this one's for you. 🧵

$HOOD down double digits after yesterday’s Q1 miss. Same for $SOFI after this morning’s Q1 beat. $COIN dropping in sympathy too🩸 Fintech’s being sold as a factor today, not on fundamentals. I’m sticking with $HOOD: net deposits +22% annualized, platform assets +39%, Gold subs +36%, event contracts +320% (nearly caught crypto in absolute dollars). Crypto rev -47% got the headline, but the wealth platform keeps compounding. Adding if it bases lower 🫡

8 hours of sleep debt 😩 I blame @NousResearch Hermes Agent! I have an addiction… every time I say it’s time to stop and relax, I get a million new ideas and I’m back on the Mac Mini 😂

Little snippet from my Hermes Agent’s Morning Briefing on markets delivered M-F at 8:45 AM ☀️

5.5 > 4.7

@Shpigford Hermes Agent is the answer

okay, done trying to make OpenClaw work for my wife (i abandoned mine a few weeks back). what should i move her to? she’s got quite a bit of memory/session/history with openclaw.

@sharbel This is why I’m still using Opus 4.7 for my Hermes Agent 😩 Hopefully they will support OpenAI API key auth directly soon - don’t want my agent suddenly breaking if limits are hit

Wait what? Codex has a small usage limit? I have been on the $200/month Claude plan for 3 months and never once hit the limit. Yet, I’ve barely used my Codex $200/month sub this week and already hit my limit. Something isn’t adding up. 🧐

@blueprint_os Oh yeah, didn’t even think about sharing the same set of tracked files across multiple agents 🤔

@kovatech_ We do something similar. 5 agents share memory files that are all tracked in git. Rollbacks saved us a few times when an agent wrote bad context that cascaded.

Track your AI agent’s (OpenClaw, Hermes Agent) workspace in git 🧠 Mine ignores everything by default with explicit git tracking for: - docs - SOUL.md and other durable context files - custom skills - scripts You get version control, change history, easy rollbacks, pull request workflow, etc.

@JohnnyNel_ Exactly 🎯

@kovatech_ treating your agent's memory like source code is genius... version control prevents drift you can't trace back

@kayintveen I agree - I’ve been using it more than Opus 4.7 the past few days, and it’s been really good!

Been hammering GPT-5.5 That model is the first one i would be ok leaving Claude for.

Hermes Agent + Discord + Obsidian is so good 👌🏼 What other essential tools or integrations am I missing? 🤔

@NousResearch Sweet! Also, that music was 🔥

Hermes Agent v0.11.0 - “The Interface Release” Full changelog below ↓

@theo But is it definitively better than Opus 4.7?

$5 per mil in, $30 per mil out. GPT-5.5 is smart. I've been using it for a bit. It's also weird, hard to wrangle, and too expensive IMO. Double the price of GPT-5.4. 20% more expensive than Opus 4.7.

So @OpenAI GPT-5.5 is now the best model for overall intelligence and performance 🤔 Tempted to switch my Hermes Agent’s main model from Opus 4.7 for a bit to see how it feels…