At Spiked AI, I build Python services for an AI product that analyzes sales meetings, producing LLM recaps and action items, retrieving past conversations, and writing back to HubSpot and Salesforce. I led a re-architecture of the asynchronous pipeline that cut turnaround from about 20 minutes to about 2 minutes.
I also shipped the product to the GCP Marketplace with role-based access control, 2FA, Cloud Armor, and secure CI/CD. I lead a distributed cross-functional team of 15 engineers and designers, and coordinate releases with founders and customers.
Earlier, as an AI Research Intern at R. V. College of Engineering, I worked on explainable deep-learning phishing detection accepted at IEEE ICOECA 2026. My projects include Artha, a real-time multi-user paper-trading dashboard, and NavaUrja, an EV-charging planner.
