On another note, i downloaded the desktop app today and also cloned the repo a couple of hours back.
The configuration was keep changing on its own. Most probably what would have been happening is desktop app would have been starting it’s own daemon and so was my cloned repo, and both of them were referencing to same system files.
So if I am changing my model in the app, it was changing in my clone repo as well.
BTW, apart from feedback for which you’ve been very welcoming about. It’s a really great project, I’m thinking of using it as orchestrator for my rest of the specialised sub agents. I’m building something that will directly integrate it in our slack, and marketing team will send vellum a message, and it will create a new page, research content, deploy it on staging, raise PR, which currently take us 3-4 days.
Thanks.
“aggregated logs to track usage”: is this aggregated logs anonymous? Does it include anything from my conversation?
Also I checked the docs, there is no litellm support, although i was able to find my way to pass through it, but still worth it to include it in the readme.
spent a 2 month streamlining codebase for the dashboard for all the modules, once the structure became uniform, made it most modular as possible extracting anything and everything.
By then only couple of things were variable on adding new modules like endpoint, column names, filter etc.
so created a frontend questionnaire for PM that will forward questionnaire as payload to this agent in the codebase and it will use that questionnaire to extract anything that is variable, then create new modules based on pre validated code, do a PR review, run UI tests, deploy it on QA, raise PR and send a slack alert.
@kawal279@kushgrwl@amishthaker_ second time costumers orders from WhatsApp -> you(restaurant ) booked the porter as soon as ready -> delivers to customers.
A couple of steps can be automated in between, but some have risk as well like who will take acrobatically accountability if food goes missing in between?
@vinodchendhil although completely unrelated, but i was lately thinking of a conversational AI that would call customers who bought recently and exclusively offer them certain discount if they buy again in certain period, what you think of it?