Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/touhi99/genai-job-agents
A LLM Agent with Langchain/Langgraph helps to analyze CV, look relevant jobs via API, and write a cover letter according to it
https://github.com/touhi99/genai-job-agents
agent langchain langgraph linkedin llm openai python resume
Last synced: about 5 hours ago
JSON representation
A LLM Agent with Langchain/Langgraph helps to analyze CV, look relevant jobs via API, and write a cover letter according to it
- Host: GitHub
- URL: https://github.com/touhi99/genai-job-agents
- Owner: touhi99
- License: mit
- Created: 2024-04-06T00:57:22.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2024-05-01T15:58:30.000Z (7 months ago)
- Last Synced: 2024-05-02T10:55:53.072Z (7 months ago)
- Topics: agent, langchain, langgraph, linkedin, llm, openai, python, resume
- Language: Python
- Homepage:
- Size: 44.9 KB
- Stars: 5
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## GenAI-job-agent
### Description
This is a simple Langgraph/Langchain based AI agent as a learning experiment of how LLM-based agent will work. Motivation is to automate better seaching in API (in this case Linkedin job), find suitable matching given user resume and write a cover letter for the most matching job. There are different usecases can be extended on top of the current design.
- Langgraph example inspired from [Notebook](https://github.com/langchain-ai/langgraph/blob/main/examples/multi_agent/agent_supervisor.ipynb?ref=blog.langchain.dev)
- Warning: Linkedin *Unofficial* API. Using it might violate LinkedIn's Terms of Service. Use it at your own risk. [Github](https://github.com/tomquirk/linkedin-api)
### Installation
```pip install -r requirements.txt```
### Usage
These environment variables are required:
```
OPENAI_API_KEY=
LINKEDIN_EMAIL=
LINKEDIN_PASS=
LANGCHAIN_API_KEY=
LANGCHAIN_TRACING_V2=true
LLM_NAME= groq/openai
```
Then run on terminal:
```streamlit run app.py```
Currently works well only with OpenAI GPT-4 / Llama/Groq still unstable.
### Agents
```
+-----------+
| __start__ |
+-----------+
*
*
*
+------------+
***| supervisor |****
******* +------------+* *******
******* *** *** *******
******* * *** *******
******* ** *** *******
**** +----------------------+ ** ****
* | supervisor_condition |** * *
* +----------------------+* ********** * *
* ***** * ***** ********** *
* ***** * ***** * ********** *
* *** * *** * ***** *
+-----------+ +---------+ +-----------+ +----------+
| Analyzer | | __end__ | | Generator | | Searcher |
+-----------+ +---------+ +-----------+ +----------+
````
### TODO
- enrich linkedin search with more params
- bug fixes: Groq Llama, unstable, due to inappropriate routing / token limit