Most coaching apps prescribe rigid schedules and punish missed sessions. Bloom-for-Learning adapts Stanford’s Motivational Interviewing coaching model into a multi-agent LLM system with a ReAct calendar tool loop, MCP calendar sync, and local-first memory — co-creating study plans instead of dictating them.
Beyond fine-tuning: LoRA with DistilBERT detected dissatisfaction in ecommerce reviews. Beyond the improved results, I found bugs and trade-offs that no tutorial tells you about.
A complete churn prediction engine with XGBoost, FastAPI, CLI, dashboard and open-source architecture for winning back at-risk customers. Applied machine learning to real business problems.
Last-click attribution can distort your entire marketing investment. I analyzed 35,000 e-commerce journeys using Markov Chains, Shapley Values, and XGBoost to prove which channels were stealing credit — and which ones were being ignored.
A predictive lead scoring system that combines ML and explainability to help B2B sales teams focus on the leads most likely to convert.