Pratik Jagtap

I'm building production AI agents at HireTalent that automate high-volume recruitment workflows across sourcing, screening, and submission.

I architect a unified AI application using LangChain, RAG, vector databases, tool calling, and multi-step reasoning, with a React frontend deployed on Vultr and PostgreSQL. I've also contributed to a human-in-the-loop onboarding agent and a YOLO-powered CCTV monitoring agent for real-time event detection and reporting.

Previously at Rubixe, I built predictive demand forecasting models for spare-parts inventory across a B2B service network. I used Python, Pandas, Scikit-learn, feature engineering, and exploratory data analysis to turn analytical findings into actionable inventory recommendations.

I'm focused on reliable agent context, prompt optimization, memory management, external API integrations, and practical AI systems that reduce manual operational work.