Hedge fund multi agent system: Build a meta-tool calling multi-agentic system using Strands SDK #32
Conversation
|
Hi @akshseh, I have made the commit to the repo to change the name of the .env template file. |
There was a problem hiding this comment.
Do we need this here? We might need an approval from legal to publish on opensource because this was generated by an LLM
There was a problem hiding this comment.
Not needed, will delete it if this requires additional approval.
There was a problem hiding this comment.
Excellent architecture, I am open to suggestions on how to make it easily readable.
At the moment it is very difficult to read on github readme.
|
|
||
| 🚨**THIS AGENTIC SYSTEM USES **[CLAUDE 4 SONNET](https://www.anthropic.com/news/claude-4), released on 05/22** | ||
|
|
||
|  |
There was a problem hiding this comment.
Can we make it architecture.png. Trying to be consistent for all samples
| 🚨**THIS AGENTIC SYSTEM USES **[CLAUDE 4 SONNET](https://www.anthropic.com/news/claude-4), released on 05/22** | ||
|
|
||
|  | ||
|
|
There was a problem hiding this comment.
Please add a table like this:
Reference: https://github.com/madhurprash/samples/tree/main/02-samples/03-aws-assistant-mcp
|
|
||
| 2. Install `uv` and Python dependencies: | ||
|
|
||
| ```bash |
There was a problem hiding this comment.
This will not tun on windows. Can we link to uv getting started guide instead.
| 4. Create a `conda` kernel. This is needed to run the notebook on your EC2 instance: | ||
|
|
||
| ```bash | ||
| uv add zmq |
There was a problem hiding this comment.
Why do we add zmq separately and not part of pyproject.toml?
This comprehensive example demonstrates a sophisticated Multi-Agent Collaboration (MAC) system designed for hedge fund analysis using the Strands Agents SDK. The system implements a hierarchical architecture with specialized financial analysis agents that work together to provide comprehensive investment insights.
View the architecture diagram below that shows how you can build a mutli-agentic system using an agent-as-tools pattern, with meta-tooling, human-in-the-loop, observability, detailed traces and logging enabled for a financial system:
🚨**THIS AGENTIC SYSTEM USES CLAUDE 4 SONNET, released on 05/22