I'm 49. I'd never written a line of code. I started learning AI on 20 April 2026.
Six weeks later I built Mission Control, and it's now live, free, and open source.
It's a personal agentic OS, not a chatbot wrapper. Chat with one agent, or hand bigger jobs to a pipeline where a team of agents works in parallel, on steps they can't wander off. Runs on your machine, your models, local or cloud. Your data stays yours, every step logged, no black boxes.
Clone it, run it, tell me what I got wrong. Link below.
@martin_valchev_ The amount of times I've don't this in the last 2 months.
I started to look at it as breaking is learning, i learned much more breaking stuff than i would have it everything worked perfectly
@mxcl If i'm reading right and you created homebrew, you have already accomplished more than most ever will.
Give yourself a break brother, you clearly already have the talent
@daicandev If it gains traction then thats an option. i built it to solve some previous issues i was having. people may not care as ive realise there are a lot of similar things out there.
I still believe mine has a lot others dont though
It's my first ever build, i didnt know what Agentic Ai was 2 months ago. so i didnt want to try and charge people.
It's an Agentic OS. The gif below shows the process
Dave sends a pipeline prompt. The orchestrator breaks it down, the workers research in parallel, and the whole run is on screen: who's working, tokens, cost, live.
You get finished reports you can open, not a chat reply.
You can also chat to the agent (Dave) at the same time from a separate instance using different models/providers.
@daicandev yeah my build is completely free and opensource, if people dont want to use but can read code, even just doing that and giving feedback would be appreciated
This is my first ever build with 0 coding experiance so just wanted feedback so i know im on the right track
@theo_jil Funnily enough that’s how I started, basically built a copy and paste of openclaw, until I realised exactly what you just said, so I took the good parts of openclaw, the good parts of Hermes, and others. Then built a version that solved my issues, not a copy
@DanielSmidstrup I don’t, this is my first ever build and it is more for the issues I was struggling with. Trying to make automation as clean and error free as possible
Didn't want to just keep posting screenshots so here is what my Agentic OS does. here is a real pipeline test, start to finish.
The orchestrator breaks it down, the workers research in parallel, and the whole run is on screen: who's working, tokens, cost, live.
You get finished reports you can open, not a chat reply.
You can also chat to the agent (Dave) at the same time from a separate instance using different modes/providers. I will post that shortly
Runs on my own machine. Free and open source.
https://t.co/w6EdvtksTg
@buildinpublic My very first build. 2 months ago i didn't know what Agentic AI was.
Now i've built my own Agentic OS
This is a running pipeline start to finish.