We did a lot of events and it is very very tedious to save info, follow and research people after meeting them. It takes a lot of time and we sometimes forget about specific info.
We built the backend with the agent orchestration in Python with LangChain and built the UI with HTML. For each integration we used the different services APIs and their authentication systems. For the linkedin tool, we used Anon. For the web search, we used Exa. And the model used is Mistral Large. Mistral Large.
Challenges we ran into:
- Authentication with agents
- Send the first linkedin message with Anon. Anon can only send message to people we already had a chat with.
Accomplishments that we're proud of: We managed to have fast and reliable agent that can achieve multiple tasks by connecting to different tools
Built With
- anon
- azure
- exa
- html5
- langchain
- microsoft
- mistral
- python
- salesforce
Log in or sign up for Devpost to join the conversation.