EngageAI: Automating Client Outreach

• 

  Project Screenshot

About the Project

• Inspiration:
  🤔 The idea came from the challenges in sales/business development within software service companies. These companies struggled with manually identifying job postings, crafting personalized outreach emails, and matching portfolios.
  💡 I aimed to automate the outreach process, making it more efficient and increasing client response rates in a competitive market.

What I Learned

• Technologies Used:
  🚀 Gained hands-on experience with Llama 3.1 (LLM), LangChain (data extraction), ChromaDB (portfolio matching), and Streamlit (UI creation).
  📊 Learned how to integrate Retrieval-Augmented Generation (RAG) for personalized email generation.
  🧠 Deepened understanding of web scraping and managing vector databases for skill-specific portfolio retrieval.

How I Built It

• Core Features:
  🔍 Scraped job portals for job descriptions and required skills.
  ✍️ Integrated LangChain for data extraction and Llama 3.1 for generating tailored emails based on job requirements.
  📂 Used ChromaDB for portfolio matching and retrieval.
  🎨 Created a user-friendly interface with Streamlit, allowing users to input job links and generate custom emails instantly.

Challenges I Faced

• Key Challenges:
  📝 Ensuring emails were both accurate and engaging enough to stand out in a competitive field. Fine-tuning Llama 3.1 prompts took several iterations.
  💾 Optimizing large-scale job portal scraping and managing the vector database for portfolio matching to ensure data consistency and system performance.

Outcome

• The project improved productivity by 25% and increased client response rates by 30%.

Built With

• azure
• chromadb
• github
• langchain
• llama-3.1
• llama-3.1-api
• python
• streamlit
• web-scraping