#general
Jul 23July 24, 2026
Hii Irfan here from India Comp Sci Grad with specialization in AI/ML and currently working as a data analytics and also exploring agents, MCP etc
Hi everyone! π
I'm Gitansh Kapoor from New Delhi, India, currently working as a Platform Engineer.
I work across a range of technologies, including Linux, Docker, Kubernetes, Backstage, Harness, GitOps, Serverless CI/CD Platforms. I'm also actively upskilling in the AI-native era, exploring MCP, Harness tooling, Agentic Automation, AIOps, and MLOps.
I joined this hackathon to build something exciting, learn from others, and connect with fellow creators. Looking forward to collaborating, building impactful products, and making some great connections along the way! π π
If anyone would like to connect, feel free to reach out. Sharing my GitHub and LinkedIn below:
GitHub: github.com/GitanshKapoor
LinkedIn: linkedin.com/in/gitansh16k
hi team!
my app uses the slack mcp and does real work in the channel it's connected to. now how do I record that for the submission? because it's just capturing the page it's recording.
can we get an option to submit our own screen recordings? @user @user
@user, after logging into archestra ai with admin credentials we have to change the default credentials ?
Hello! I am Manul from India, working as a backend engineer since past 3 years. I have shipped RAG chatbots and developed distributed backend services at my organisation in insurance domain. I have worked in: TypeScript, Nest.js, Python. Exctied to be a part of the community and contribute!
π2
1 reply
Good morning everyone, how's your progress? π
I am facing rate_limit error , Too many requests. Please wait a moment and try again.
any quick workaround ?
Status: 429
Type: unknown_api_error
Message: Failed after 3 attempts. Last error: 429 status code (no body)
Raw Error:
{
"url": "<http://localhost:9000/v1/cerebras/0ed72c90-cea3-490b-ac3c-d04843dc2fb0/chat/completions>",
"statusCode": 429,
"isRetryable": true,
"responseBody": {
"error": {
"message": "429 status code (no body)",
"type": "unknown_api_error"
}
}
}```
Hey all, weβre having a Q&A webinar for the Hackathon participants today.
It starts in 3 hours - weβll show you hot to build an App in Archestra + answer any questions you have.
π Register here
Hey! Where is the Apps Gallery repo or form to submit my hackathon project as suggested?
May I ask someone to submit a toy app to test the submission flow? It won't impact the final submission π
For those who have their app using slack mcp.did you encounter the redirect_uri did not match any configured URIs error. @user
Weβre starting our Hackathon intro event in 10 minutes! Join here
hi guys. Can I track costs per user in archestra (without making a team or proxy per user π ) ?
I tried submitting my project and got this
Hi team,
I'm getting an error while installing
microsoft__playwright-mcp. Can anyone help me with a solution?
My Windows WSL is broken, so Docker Desktop won't start. Is there a supported way to run Archestra using GitHub Codespaces or another cloud environment for the hackathon?
hi could anyone submit via the βsign in via githubβ option?
Read-only live mirror of Archestra.AI Slack
πJoin the discussion withAI enthusiasts!hi guys. Can I track costs per user in archestra (without making a team or proxy per user π ) ?
8 replies
Yes! You can use *tags* on API calls to track costs per user. Add a
tags array to your requests: ``json { "tags": ["user:alice", "project:demo"], ... } `` Then filter by tags in your usage dashboard or via the API. Docs: https://archestra.ai/docs/platform-quickstart (see the tags section) π€ Slack Agentπ1
Thanks! π Let me know if you need anything else! π€ Slack Agent
but to make it more precise: where do I these tags? When Claude is making the requests to the llm proxy I cannot enrich the request with custom tags can i ?
Good point! If Claude (or another model) is making requests through the proxy, you can't inject tags at call time. *Options: 1. Proxy-level default tags β set tags on the proxy config itself (e.g.,
"defaultTags": ["user:alice"]). You'd still need one proxy per user though. 2. API key tags β create per-user API keys with tags baked in. Usage gets attributed via the key. 3. Custom header* β some setups let you pass X-Archestra-Tags header if the client allows custom headers. What's your setup β are you using MCP, or direct API calls from your app? π€ Slack AgentIf you're using claude, Connect page should generate per-user keys so your logs should be attributed
ok
Got it! So with Claude via the Connect page, your logs are already attributed per user. Let me know if you need help with anything else! π€ Slack Agent