EDIT: WE HAD OVER 100 PEOPLE SIGN UP IN TEN MINUTES....GOING TO BE ROLLING OUT ACCESS ON SLOWER BASIS NOW SO I DONT RUIN MY BANK ACCOUNT WITH TOKEN SPEND FOR YALL (im covering ai costs for everyone) but keep applying we are keeping track and will be letting more people in and reaching out about getting you all access one by one especially VIP. Our Platform is here, the thing a lot of you have been asking for and I have been teasing Take your ICM folders and second brains, put them in the cloud, and work on them together in real time. You upload your folders and files. One strong model with a good harness reads the map and becomes the agent you need. No new agents to build, no zip files to pass around. Each workspace runs in its own container. It renders your markdown, edits your files, installs packages, runs Python and Playwright, and multiple people can work in the same space at once. No Mac Mini needed to keep your data separate either. We're opening it to VIP and alpha testers first, and I'm covering the AI cost while we test so cant do a full role out yet. Free AI use which is great for those users, not great for my bank account haha. ✍️ Sign up for access: https://docs.google.com/forms/d/e/1FAIpQLSdSnToxclxt8EgoB2GizrEvHP7gzOedVcRXdCkpi1SZwR8ZfA/viewform 🔗 Questions, or want to work heavier with this? Email info@eduba.io

This is the only place I can talk local ai without getting blank stares, so let's actually see what everyone is running. I'll go first. here's the whole stack: HARDWARE ▸ Alita - personal dual-3090 rig at home. 48gb total vram, 128gb ram. kept totally separate from anything work related. MODELS ▸ primary: qwen3.6-35b-a3b on gpu0, served by llama.cpp. mixture-of-experts, ~3b params fire per token. q8 kv cache, 256k context. ▸ router: a ~9b qwen on ollama out front, deciding what's actually worth waking the big model for. ▸ embeddings: qwen3-embedding 0.6b and nomic-embed-text, same ollama instance. AGENT LAYER ▸ hermes, running as Alita here. three daily jobs: morning brief, work email summary, personal email summary. all land in discord. ▸ weekly maintenance runs on a different scheduler with hermes as the watchdog, on purpose. runner dies, watcher yells. watcher dies, the daily jobs go quiet and I notice. two failure domains instead of one. MEMORY ▸ obsidian vault as the source of truth. ▸ hybrid search over it - bm25 plus local embeddings, running in-process. no extra service to babysit. ▸ qdrant for the vector-store work that actually needs one. INTERFACE ▸ the observatory - a front end I built over the vault. not a dashboard. a command center I actually work from: projects as boards, sessions as a timeline, drafts moving through a publishing pipeline, all of it reading and writing the same markdown files. The one piece I would not give up is the vault. the models change every few months. the notes are forever. The thing that taught me the most had nothing to do with hardware. I did kill a pile of containers nothing was consuming - n8n, a web ui, a couple databases - but that was housekeeping. the real one was giving up on folders. I spent a long time sorting notes into the right directory and it was wrong again six months later. Now every note carries its own frontmatter and the folders are just a cache I regenerate from it. Type isn't a folder. Status isn't a folder. The structure lives in the file.

Context: My company is hell-bent in still making excel work for everything. 10-15 worksheets inside one file, THOUSANDS of rows and columns per sheet. Curious question, do agents follow the same ICM principles when you point them at an excel file? Ex. Routing logic in first worksheet so the agent knows where to go instead of searching the entire workbook for something.