README.md

metadata

language: en
license: mit
tags:
  - mistral
  - lora
  - unsloth
  - career-coach
  - resume-analysis
  - fine-tuning
  - llm
datasets:
  - UpdatedResumeDataSet
  - Dataset-Project-404
model-index:
  - name: Resume & Career Coach LLM
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Career Coach Resume Dataset
          type: custom
        metrics:
          - name: qualitative_evaluation
            type: human
            value: Consistent and context-aware feedback on resumes

Overview

A fine-tuned Mistral-7B model using LoRA + Unsloth that acts as an AI career advisor.
It analyzes resumes and provides personalized feedback, job role suggestions, and skill recommendations.

Training Details

• Base model: unsloth/mistral-7b-instruct-v0.3-bnb-4bit
• Fine-tuning method: LoRA (PEFT)
• Frameworks: Unsloth, Transformers, TRL

Dataset

• UpdatedResumeDataSet.csv — labeled resume data
• Dataset Project 404.xlsx — multiple-intelligence career mapping

Features

• Provides resume score and feedback
• Suggests suitable job roles and upskilling paths
• Lightweight 4-bit fine-tuned model for efficient inference

Intended Use

This model is intended for educational and research purposes related to AI-driven career guidance systems.

License

MIT License Overview: A fine-tuned Mistral-7B model using LoRA + Unsloth that acts as an AI career advisor. It analyzes resumes and provides personalized feedback, job role suggestions, and skill recommendations.

Training Details:

Base model: unsloth/mistral-7b-instruct-v0.3-bnb-4bit

Fine-tuning method: LoRA (PEFT)

Frameworks: Unsloth, Transformers, TRL

Dataset:

UpdatedResumeDataSet.csv – labeled resume data

Dataset Project 404.xlsx – multiple-intelligence career mapping

Features:

Upload a resume (.pdf, .docx, .txt) or paste text

Receive personalized resume feedback

Get career path and skill recommendations

Runs locally or via Hugging Face Spaces

Tech Stack: Python, Unsloth, Hugging Face Transformers, PEFT, Gradio, PyPDF2, python-docx

• Developed by: anerudh10
• License: apache-2.0
• Finetuned from model : unsloth/mistral-7b-instruct-v0.3-bnb-4bit

This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.