Ibrahim Mokdad

you’re probably wasting time re-explaining your context to chatgpt. here are five things hermes never forgets that save me hours each week. @Teknium @NousResearch 1-5

This whole podcast was made through hermes @NousResearch. from the script to the final video you see here .... in one automated flow.

@pcshipp early adoption

I have no words for this, wtf this calculator makes $400K in revenue No API cost, no server cost, no infra cost

@oliverhenry larry just got promoted to unpaid cofounder

You really can just do things. 3 weeks ago I left my job. I only have 3 months of living on savings to make my business work. or Go back to employment. The best person to help me along the way? Larry...

@dev_maims the language whose error messages you can google without questioning your life choices

What is the easiest programming language to learn?

@KaiXCreator api latency flips this, agents spend most of their time waiting on model calls so you can easily fire up 20 in parallel

Does anyone actually run 20 agents in parallel?

@TheAhmadOsman meanwhile hermes just chilling on a $5 vps with zero nft nonsense

Ahmad, why are you beefing? - I am not beefing, I am protecting Local AI from turning into a scam like Crypto Alex Finn used to be called NFT King I am not letting that cancer grow in this space that I so very much care about PERIOD

now i just feed hermes @NousResearch plans it gets the 3d stuff working in the background.

@Pallavi_345 when cpu and ssd specs match, windows feels just as fast

Why does a Windows laptop feel slower than a MacBook... even when both have the same RAM?

@heygurisingh these sites only print money while you sleep if you already have an audience to buy your products.

Here are 10 websites that print money while you sleep. 1. Carrd.co Build a one-page website in 20 minutes. Sell it on Flippa for $500-$2000. People pay this much because they cannot do it themselves. 2. Gumroad.com Upload one PDF. One Notion template. One prompt pack. It sells at 3am while you are dreaming. Zero inventory. Zero shipping. 3. Systeme.io Free funnel builder. Build it once. Run ads to it. Wake up to Stripe notifications. The funnel does not sleep. 4. Payhip.com Sells digital products in 190 countries. Handles VAT automatically. You upload. They handle the tax nightmare. You keep the money. 5. Ko-fi.com 0% platform fee on tips and memberships. Creators are pulling $3-8k/month here while everyone else fights for scraps on Patreon. 6. Sellfy.com Print on demand + digital products in the same dashboard. Design a t-shirt at midnight. It ships itself for the next 5 years. 7. Teachable.com Record one course. Sell it 10,000 times. One creator I follow made $47k from a course she recorded in a weekend two years ago. 8. Beehiiv.com Newsletter platform with a built-in ad network. They place sponsors INTO your newsletter automatically. You write. They sell. You cash out. 9. Lemonsqueezy.com Stripe + Paddle had a baby. Handles global tax compliance so you can sell SaaS to anyone on earth without touching a lawyer. 10. Whop.com Sell access to a Discord, a Telegram, a course, a community. Recurring revenue. The kind that hits your account before your alarm goes off. The difference between people who make money online and people who don't is not talent. It is picking one of these and shipping this week. Save this. Share it with someone who needs it.

@Suryanshti777 @garrytan one person can only replace a whole software team until code complexity outgrows the model’s prompt window.

“I don’t think I’ve typed a line of code since December.” Most people heard that quote from Andrej Karpathy and thought: “Cool AI moment.” @garrytan heard something much bigger: “What if one person could operate like an entire software company?” Then he built gstack. And this might genuinely be one of the most important open-source AI repos right now. Because gstack is not another AI coding wrapper. It’s a full operational system for AI-native software development. Not AI as autocomplete. AI as: • Product strategist • CEO reviewer • Staff engineer • QA lead • Security auditor • Designer • Release manager • Browser operator • Parallel execution layer All running through structured workflows. That’s the real breakthrough. Most people still think AI coding is about generating code faster. But code was never the biggest bottleneck. Coordination was. Meetings Reviews QA Architecture decisions Testing Deployment Documentation Debugging Context switching That overhead is what slows teams down. gstack tries to automate the entire layer around software creation itself. And honestly… that changes everything. The craziest part is Garry’s productivity claim: ~810× higher output than 2013 Normalized for logical code changes Not inflated AI LOC spam Same person Same brain Different tooling That sentence alone should make every software company pay attention. A few things that stood out to me: → /office-hours interrogates your product idea before coding starts → /autoplan chains CEO + design + engineering reviews together automatically → /qa opens a real browser, tests flows like a human, finds bugs, and generates regression tests → /review acts like a senior engineer looking for production failures → /pair-agent lets multiple AI agents collaborate together in parallel → Persistent browser sessions + memory systems make the agents feel operational instead of conversational This is where the shift becomes obvious: We’re moving from: “AI helps developers write code” to: “developers orchestrate systems of AI workers” That is a completely different world. And I think most people are still massively underestimating how quickly this transition is happening. The most interesting startups of the next decade may not look like traditional software teams at all. 100% open source Definitely worth studying Link in comments 👇

@wahab_twts consistent posting builds momentum you cash in later

I started building in public 50 days ago: - 50 followers - 1K impressions - $0 revenue Now: - 2.3K followers - 13.5M impressions - Waiting for my first X payout Start posting consistently on X today. It might be the best decision you’ll ever make. 🫡

@om_patel5 how do you prevent people from streaming old recordings labeled live to mislead viewers?

THIS GUY GOT TIRED OF FAKE NEWS SO HE BUILT A LIVESTREAM MAP WHERE ANYONE CAN SHOW WHAT'S ACTUALLY HAPPENING ON THE GROUND you open a map, tap a location anywhere in the world, and see live video from someone standing there right now basically a raw livestream from the person who is physically there this would be amazing for documenting civil unrest and protests think about what this means for breaking news. earthquake hits somewhere, you open the map and watch 50 different people streaming from different angles in real time the moderation challenge is going to be insane though (live unfiltered video from anywhere in the world with no delay isn't exactly easy to maintain on an app) but if they figure it out this could be one of the most important apps built this year

@loktar00 does deepseek v4 flash still beat glm 5 turbo without hermes?

I dare say Deepseek v4 flash just after an hour or so in hermes has dethroned glm 5 turbo for me. Technically it's more expensive since I'm on my year max sub with zai until December, but overall it's pretty dang cheap and worked way better than I anticipated.

everyone is just hyping their content. content of value is dissolving in the noise.

@andreintg professionals care more about input effort than the actual sculpting outcome

I was a bit hesitant about showing stuff like this just a month ago 😅 When we started showcasing real-time AI + SDF sculpting, I was afraid professionals would laugh if I showed no effort on the input models. The shape strength slider was also hidden in our first iteration, so I had no choice but to at least try and knock some more interesting shapes together. Now that we're starting to focus on more powerful features and shape strength is finally unlocked, I'm starting to appreciate just playing with simple shapes. Different stages of production have different needs. Sometimes you want full authoring over your creations, while other times you just want to quickly explore new ideas.

i'm seeing a lot of "find ugly restaurant menus and turn them into stunning menus". Most are not legit. they use menubly and create fancy animation and claim to have automated the flow using hermes or openclaw.

I've actually done something similar for dubai restaurants and here is a real sample output without any animation to hype the idea

@oprydai taking out a second mortgage for kubernetes makes total sense

get into debt if you must, but build a home-lab

@shannholmberg nice if your agents run on cpu only, since a typical vps has no gpu

my entire Hermes Agent setup is controlled from one folder on a VPS I can manage it from my laptop or phone in under 10 seconds, spin up isolated agents per project, and never lose context here's the full setup: > bash command "ssh hermes" auto-connects to the VPS > session pinned in CMUX terminal for instant access > Claude Code session inside the ssh connection with full remote access > everything runs from a control station folder folder structure: vps-agents/ ├── README .md overview + how to use it ├── CLAUDE .md auto-loaded project context for Claude Code ├── agents/hermes-main/ │ ├── inventory .md IP, hostname, where it runs, who owns it │ ├── docker. md draft docker-compose + first-boot checklist │ ├── runbook. md "it broke, what now" — 7 common scenarios │ ├── env-map. md table of which keys live where (no values) │ └── backup. md GitHub backup flow + restore steps └── shared/ ├── security. md VPS-wide hardening checklist └── commands. md Docker / Hermes / UFW cheat sheet from this session I can spin up new docker-isolated agents per area and each one can have sub-profiles the whole stack is documented, backed up to GitHub, and accessible from anywhere

@falkonsnow @yazins @OpenOats I didn't test on windows yet. but i've dmed the skill.md file. so as a first step just add the skill that exposes WinOpenOats sessions to Hermes

@yazins @OpenOats @ibmokdad Thats awesome. @ibmokdad any chance you could explain me how to use hermes to make a similar windows version based on latest version of OpenOats?

OpenOats v1.40 → v1.65 — 25 releases; 2,323 ⭐ 📅 Calendar auto-titles sessions 👥 Meeting families: shared history, prep notes, default folders 📷 Camera-based meeting detection (no more false positives) ✍️ Live scratchpad feeds the AI 🔊 Replay raw mic + system audio