I am Aman Raj, a Full Stack AI Developer and Software Development Engineer II with 2+ years of experience building scalable APIs, event-driven systems, developer tooling, and AI-powered workflows. I'm proficient in Python, FastAPI, Django, Node.js, and cloud-native backend services, with hands-on exposure to LLM integrations, function calling, retrieval-based workflows, and production-grade observability. I have a strong track record delivering reliable data-intensive systems, microservices, and automation that improve user experience, data access, and platform performance.
Migrated analytics APIs from Apache Pinot to StarRocks, redesigning schema structures and indexing to reduce query latency by 30%. Designed high-throughput real-time alerting workflows using AWS Kinesis, Apache Flink, API Gateway, and AWS Lambda, reducing production errors by 20%. Engineered a highly available notification/email microservice using Amazon SQS and SES to process millions of event-driven alerts, integrating SSE to auto-triage 2% anomalous scans before customer impact.
Developed backend integrations and middleware wrappers with OpenAI function calling to create structured, secure API layers for LLM-powered workflows, reducing manual processing and improving response quality.
Designed scalable API architectures and database schemas for 3 production web platforms, prioritizing performance and modular code.
Designed and expanded automated testing suites (unit, integration, and UI testing) using pytest and Jest, enhancing system reliability by 6%. Architected an offline-first synchronization engine and database caching layer using Gzip for local storage, reducing database size by 6x and increasing active mobile contributions by 15%.
Real-time WebSocket-based code editor synchronizing concurrent text editor states, boosting real-time collaboration efficiency by 30%. CI/CD via GitHub Actions running Jest test suites with automated deployment to Azure.
View Project
High-performance URL shortener built with Node.js and Express supporting high-throughput URL generation and redirection. Integrated Redis caching alongside MongoDB, reducing average database lookup times by 90%.
View Project
A Social media app made with Flutter and using Firebase as backend, being coupled with some good features it gives users a rich experience without any ads.
View Project
A web application to reduce the food wastage by listing the food on website before it goes bad. Made with HTML and CSS and Php using as backend.
View Project
An application made with Flutter that let's user chat with fellow users, featuring end-to-end encryption through assymetric encryption.
View Project