TRANSMISSION_INIT_00
Jyotishman
Shandilya
Backend engineer at Outcomes®, working on systems that scale. I also teach software engineering publicly on @jyoti_codes — short-form content that makes the hard stuff actually make sense. No gatekeeping, just good explanations.
WORK_HISTORY
Delivered SMS engagement campaigns for Walgreens reaching 120,000+ patients through rules engine work and end-to-end validation. Built a generic receiver API to unify Twilio and Sinch delivery reports, and shipped a Spring Batch replay service that recovered 100,000+ messages during infrastructure outages.
Productionised an AI classification system on Amazon Bedrock that Walgreens uses to classify inbound messages from 1M+ patients. Improved model accuracy by 85% via a reinforced LLM training pipeline, cut manual review effort by 80%, and wired output into SQS and SFTP pipelines for automated downstream actions.
Built a GenAI-driven RAG chatbot for lab automation that cut response times by 50%. Optimised processing with a GPU-accelerated library (30% faster execution) and automated software installation via batch scripting — reducing environment setup time by 92%.
Built interfaces for cybersecurity products — responsive, fast, and designed to not confuse the people using them.
Led the web development wing, mentored juniors getting into full-stack development, and shipped internal projects. One of the best decisions I made in college.