I’m currently building a series of Minimum Viable Product (MVP) projects to gain hands-on experience with AI including exploring algorithms, learning the ecosystem, techniques for evaluation & benchmarking, optimization & fine-tuning, automation & orchestration, and AI Agents. While I don’t consider myself an AI developer, embracing Vibe Coding has helped me understand the flow of AI systems and system constraints. My goal is to deepen my understanding of AI’s nuances and sharpen my approach to building AI-enabled products end-to-end; from use case discovery to deployment and automation. These experiments have helped me to create a mental model on how various models work, how they can be adapted for specific use cases and how the pre-training/post-training shapes their behavior.

1️⃣ Diabetes Predictor


2️⃣ Internal Document Recommender


3️⃣ Anomaly Detection in Expense Claims


4️⃣ Reasoning Model