Skip to content

anshumaantgi/TPRS

Repository files navigation

TPRS 🤖

Overview 💬

TPRS is a sophisticated system designed to enhance career development by extracting skills from users' resumes and aspirations, defining SMART goals, and recommending tailored training plans. Leveraging advanced multi-stage LLM ReAct agents, TPRS provides a comprehensive approach to skill development and goal achievement.

Key Features ✴️

Skill Extraction: Automatically extracts skills from users' resumes. SMART Goal Extraction: Identifies SMART goals from user aspirations. Training Plan Recommendation: Suggests a personalized training plan to achieve the defined SMART goals. Natural Language Modification through ChatBot: Allows users to edit the training plan, aspirations, and skills using a chatbot interface.

Technologies Used 💻

  • Programming Languages & Frameworks**: Python, Flask, Streamlit
  • Large Language Models: Claude-3-Haiku, Mistral 8x7B
  • Libraries: LangChain, LangGraph ReAct Agents
  • Cloud Services: AWS Bedrock (Prompt Management System, KnowledgeBase for RAG), AWS S3

Architecture 📊

Overall Architecture

Overall Architecture

Skills Extraction Agent Architecture

Skills Extraction Agent Architecture

SMART Goal Extraction Agent Architecture

SMART Goal Extraction Agent Architecture

Training Plan Recommendation Agent Architecture

Training Plan Recommendation Agent Architecture

Product Demonstration 📺

Installation ⬇️

  • Install the required packages by running the following command:
pip install -r requirements.txt
  • Run the following command to start backend server:
python backend.py
  • Run the following command to start frontend server:
streamlit run frontend.py

Note: AWS Bedrock credentials are required for full functionality.

Usage

  • Open the frontend server.
  • Enter your resume and aspirations.
  • Click the submit button to receive your personalized training plan recommendation.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors