Aasthik User

At Usenseon Labs, I'm building and deploying an end-to-end healthcare platform that connects ESP32 devices, real-time AWS backend services, and a mobile application for low-latency production data processing.

I work directly with the founder to prototype, test, and deploy features from user feedback, improving performance across embedded hardware, backend services, and mobile applications.

I've built Anamnesis, an open-source local-first AI code assistant that preserves Git history, architectural decisions, debugging knowledge, and code changes in a graph-based memory system. I also built CareerPath AI, an agentic RAG platform using Amazon Bedrock, AWS Lambda, API Gateway, S3, and Next.js to provide explainable career recommendations.

My ML work includes a PyTorch Transformer speech normalization system trained on 2,172 recordings across 214 native language groups, achieving a final test MSE of 0.0022. I enjoy building practical AI systems, and I've been recognized as a state hackathon winner, national hackathon finalist, and speaker on ML-based pneumonia diagnosis.