koool kal

• Led the design of an end-to-end AI pipeline that automates weekly content strategy by scraping competitor data, analyzing it with a cost-optimizing multi-agent system, and auto-generating a ready-to-publish content calendar. • Co-developed an advanced multi-agent AI chatbot for expert-level football analysis, architecting the system with LangGraph, a supervisor agent, and an end-to-end RAG pipeline over YouTube transcripts. • Designed and developed a comprehensive Power BI dashboard to monitor key business metrics (user acquisition, engagement, monetization), enabling data-driven product decisions.

- Built end-to-end Reddit sentiment analysis and topic classification pipelines, with results visualized in Power BI. - Created a Discord bot with integrated Power BI dashboards to monitor community growth and engagement. - Scaled and automated outreach to 50k+ creators via a Python suite, doubling DM response rates using AI-personalized messaging. - Designed an AI agent automating creator contact info discovery & personalized DM generation, cutting manual research time. - Developed a scalable wrapper for the Google Gemini API with key cycling, throttling, token estimation, and usage tracking to optimize LLM costs. - Produced Power BI dashboards for competitor analysis & manual outreach tracking, directly informing leadership decisions and team strategies.

- Developed a real-time data pipeline using Kafka to stream data from MongoDB to PostgreSQL, ensuring a seamless data flow to live Power BI dashboards. - Conducted Survival Analysis on lung cancer patients, building a predictive model in R

- Developed an ETL pipeline in Python to retrieve, preprocess, and clean data from an Oracle database. - Trained a machine learning model to predict employee churn and deployed it in a web application with a visualization dashboard.