🤖 LinkedIn Automated Content System — built and tested end-to-end Wanted to share a workflow I put together (n8n) for automating my LinkedIn content, niched down specifically to hospitality (hotels) and hospital/healthcare automation topics. How it works: 1. I send a topic to my Telegram bot (or trigger it manually for instant posts) 2. It filters out anything that's not hospitality/hospital-related — keeps the content strictly on-niche 3. Pulls real research/facts on the topic instead of hallucinating stats 4. Writes the post in my voice using Gemini 5. Generates a matching image 6. Sends me the preview on Telegram with two real buttons: ✅ Approve & Publish or ✏️ Request Changes — no typing needed 7. If I approve, it publishes straight to LinkedIn and confirms back on Telegram 8. If I ask for changes, it loops back for a revision A few things I learned building this: - Telegram's sendAndWait node with button-based approval is way cleaner than free-text replies - Photo captions on Telegram have a 1024-char limit — learned that the hard way - Free image generation (Pollinations.ai) works great for this, no OpenAI billing needed - Keeping the topic filter simple (a keyword regex) does 90% of the job for keeping content on-niche Happy to answer questions if anyone's building something similar!

4 days of wrestling with Meta's API, and it's finally alive. I built a system that takes ONE post and automatically cross-posts it — with real photos and videos — to LinkedIn, Facebook, and Instagram. No copy-pasting, no manual uploads, no "wait let me resize this for each platform." The automation part? Honestly the easy 20%. The other 80% was untangling Facebook Page linking, Instagram's new Business Login flow, permission scopes Meta renamed mid-setup, and figuring out why Instagram was rejecting my video (turns out it needs time to actually process media before publishing — nobody tells you that upfront 😅). It's running now — LinkedIn ✅ Facebook ✅ Instagram ✅. YouTube is next, still wiring that piece in. If you're deep in Meta API hell right now — I've probably hit the same wall. Happy to share what worked.

Hey I know accelerator is a great program but I want to know can I get first clients without it?

I'm trying to figure out the cleanest way to handle two parts of my offer. One, if a new lead hits the client CRM, what's the simplest and fastest way to get that lead into their specific GoHighLevel sub-account, or is there a better way than Make or Zapier? And two, how do you give the AI access to the client's real-time calendar availability, outside of GoHighLevel, so it books accurately without double bookings? For folks who've built something similar, what's the cleanest setup? Thank you

Curious how people here are actually building their personal AI OS / agentic harness 🙂 My current setup is roughly: Local Codex instance + Notion as my operational AI brain (+ distilled into Obsidian) + Google Drive, with my important local folders/files synced to Drive in real time + GitHub (although quite poorly updated) On the VPS I run another Codex instance + Hermes. I’ve spent quite some time designing, organizing and populating my whole digital environment. At this point, my Notion is almost like an up-to-date digitalized version of my consciousness — projects, decisions, knowledge, ideas, systems and context — and Codex continuously keeps it updated and operational daily. Codex is currently my main go-to environment, while I still use Claude and Hermes separately. The next step I’m trying to solve is how to superpower all of this into one coherent system where Codex, Claude and Hermes work together, share context, hand tasks between each other and operate across both my Mac + VPS from one environment. For those building something similar: what does your architecture look like? Especially interested in how you handle: - local ↔ VPS control - shared memory/context - multiple agents/models - orchestration/routing - one unified interface vs separate tools Would love to see real setups, not just tool lists. 🙏