Sorry, unable to follow Akshay Kumaran. Please try again.
I’m a Software Engineer passionate about building scalable backend systems and AI-driven applications that solve real problems. I recently completed my M.S. in Applied Data Analytics at Boston University, where I worked deeply with Python, FastAPI, PostgreSQL, and modern cloud-native development. Through internships and projects, I’ve built production-style APIs, ETL pipelines, and multi-tenant systems with strong focus on performance, reliability, and clean architecture. My work ranges from AI-assisted developer tools and RAG-based workflows to secure data platforms using Row-Level Security and real-time processing. I enjoy working across the full development lifecycle designing systems, writing reliable backend logic, optimizing databases, and collaborating in Agile teams. Alongside engineering, I’ve mentored graduate students as a Teaching Assistant, strengthening my ability to explain complex technical ideas clearly. I’m currently actively seeking Software Engineer / Backend / AI Engineer opportunities where I can contribute, learn fast, and build impactful systems at scale.
Microsoft Power BI, Azure Databricks and +2 skills
TypeScript, Next.js and +2 skills
1. Assist faculty in evaluating student projects and assignments involving data mining, R, Python, and SQL. 2. Manage course content, assessments, and communication using Blackboard and related academic technologies. 3. Collaborate with the Instructor of Record to deliver instruction, training, and applied analytics support throughout the semester
1. Built enterprise-grade data extraction microservices through Flask-RESTX APIs and PostgreSQL databases using DLT (Data Load Tool) and OAuth2 authentication, achieving 2,500 events per calendar extraction with 30-day time range filtering, reducing manual data collection from hours to minutes and enabling automated extraction for Google Calendar and Jira Board data across multiple organizations. 2. Implemented scalable ETL pipelines through DLT framework with checkpoint-based resumption using async processing and 5-calendar checkpoint intervals, delivering pagination support for 250+ calendars per scan with 2,500 events per page, enabling resumable extractions that handle large datasets without data loss and support 5 concurrent scans with 24-hour timeout windows. 3. Designed robust API architecture through Flask-RESTX with comprehensive error handling and rate limiting using Prometheus metrics and structured JSON logging, achieving 30-second request timeouts with 3 retry attempts, supporting 1,000 requests per hour rate limits and reducing integration complexity through auto-generated Swagger documentation with 15+ API endpoints. 4. Developed production-ready containerized services through Docker multi-stage builds and PostgreSQL connection pooling using 10-connection pool size with 20 max overflow, scaling to handle 100+ records per batch with 7-day cleanup cycles, reducing deployment complexity through docker-compose orchestration with 6 services (app, postgres, redis, adminer, loki, promtail) and enabling multi-environment support (dev, staging, prod).