cvpilot¶

    AI copilot for CV/resume evaluations like a bird on LinkedIn
    — just trying to land the perfect job without ruffling too many feathers

CV Assistant UI

Basic App¶

How to setup ?¶

    Basic App uses Google's Gemini as it has good free tier options https://ai.google.dev/gemini-api/docs/rate-limits#free-tier

    Step1: Generate API key from https://aistudio.google.com/app/apikey
            and set it in .env

    Step2: Pick a model from https://ai.google.dev/gemini-api/docs/rate-limits#free-tier
            and configure it at basic_app/gemini_agent.py

    Step3: Install required libraries
            If you have virtual env then make sure to use "poetry env use"
            poetry install

    Step4: Test your code
            python basic_app/main.py

Basic App flow¶

[User Uploads Resume PDF/DOCX]
        ↓
[Resume Extraction]
        ↓
[JD Extraction]
        ↓
[Agent (Gemini / LangChain / CrewAI)]
        ↓
- Loads JD
- Compares skills, roles, certs, etc.
- Summarizes match/gaps
        ↓
[Response Returned to User]

Full App¶

    Unlike basic_app, this has a small frontend driven by streamlit

How to build ?¶

    docker build . -t cvpilot-app
    docker run -v /Users/sheelava/msashishgit/cvpilot/jd:jd -v /Users/sheelava/msashishgit/cvpilot/cv:cv cvpilot-app

How to use ?¶

    poetry run streamlit run app/main.py

Tech Stack¶

Feature	Tool
Resume Parsing	`pdfplumber`, `python-docx`, `unstructured`
LLM	Gemini LLM /OpenAI GPT / Vertex AI (GCP)
Agentic Framework	`LangChain` or `CrewAI`
Optionally Deploy UI	Streamlit / Flask

Extras¶

Using an already created virtualenv within poetry¶

Create a virtualenv inside your repo directory python -m venv .venv-cvpilot ...or... pyenv virtualenv 3.11.2 .venv-cvpilot
(the trick) Make poetry use an existing virtual environment - Setup in-project configuration poetry config virtualenvs.in-project true - Activate virtual env source .venv-cvpilot/bin/activate - Tell poetry to use existing virtual environment poetry env use /Users/sheelava/msashishgit/cvpilot/.venv-cvpilot/bin/python

About streamlit¶

    streamlit helps convert raw data into visual delight - tables, histogram etc
            https://docs.streamlit.io/
            https://docs.streamlit.io/get-started/tutorials/create-an-app

    streamlit run pick.py