Data Science & AI, ML,DL Hub

12/09/2026

🚀 Another Project Completed & Shipped

As part of Homework 2: Build and Ship an AI-Assisted Full-Stack App, I built FairShare — AI-Assisted Expense Splitter.

FairShare is a full-stack application designed to make sharing expenses easier by allowing users to record expenses, calculate balances, and determine settlements.

💻 Built with:
• React + TypeScript + Vite
• FastAPI + Python
• SQLAlchemy + SQLite
• REST API
• Pytest
• Git & GitHub

The project was a great hands-on exercise in going through the complete development workflow: planning → coding → AI-assisted development → API integration → testing → debugging → Git/GitHub → shipping.

🔗 GitHub: https://github.com/jcdumlao14/fairshare-expense-splitter

Homework completed. Project shipped. Another step forward. 🚀

ai-dev-tools-zoomcamp-homework-2. Contribute to jcdumlao14/fairshare-expense-splitter development by creating an account on GitHub.

08/09/2026

“Strong AI is built on strong fundamentals.”

08/09/2026

“A model can be accurate and still be wrong for the problem.”

08/09/2026

“When patterns become clear, possibilities become predictable.”

05/09/2026

🚀 Project Attempt 3 — LLM Zoomcamp 2026

Intelligent Financial Research with Hybrid RAG 📊🤖

I’m excited to share FinInsight-AI, an end-to-end financial research assistant built using Hybrid RAG.

It combines Vector Search + BM25 + RRF + Query Rewriting + Cross-Encoder Reranking + Gemini 2.5 Flash, with a Streamlit interface, Prefect ingestion, monitoring, and Docker.

🔗 GitHub Repository: https://github.com/jcdumlao14/FinInsight-AI

The goal: faster financial research with answers grounded in source documents. 📚

🙏 Special thanks to Alexey Grigorev, Founder of DataTalks.Club, and the DataTalksClub LLM Zoomcamp for the learning framework and practical foundation.

Another project completed, another step forward in my AI/LLM journey! 🚀

Project Attempt 3 in LLM Zoomcamp 2026, Intelligent Financial Research with Hybrid RAG - jcdumlao14/FinInsight-AI

02/09/2026

“Don’t fear complex data. Learn to ask better questions.”

02/09/2026

“Beautiful insights often begin with ugly data.”

27/08/2026

🔍 Discover hidden patterns without labeled data! K-Means groups similar data points into clusters for segmentation, pattern discovery, and smarter decisions.

27/08/2026

🚀 Build better predictions step by step with Gradient Boosting! Each new model learns from previous errors to capture complex patterns and improve performance.

27/08/2026

🌲🌲 Combine many decision trees to build stronger predictions. Random Forest improves stability and accuracy for classification and regression tasks.