SIDHARTHA VYAS

I’m a Backend & GenAI Engineer with 1.5+ years of production experience building scalable REST APIs, distributed systems, real-time processing pipelines, and AI-powered applications. I focus on turning system design into reliable, measurable performance in the real world.

In my production role, I architected MPNC, a backend platform for real-time inventory lifecycle management, batch tracking, expiry automation, and demand forecasting across multi-location operations. I designed and deployed 20+ REST APIs using FastAPI and Python, including FIFO ledger logic, automated state-transition engines, and procurement forecasting modules.

I also built performance-critical data paths—engineering SQL Server query optimization pipelines with stored procedures, indexing, and reporting modules to power operational analytics. Earlier, as a backend intern, I created secure, high-throughput REST APIs in Node.js for a logistics CRM, and optimized PostgreSQL schemas and query plans to reduce latency and improve throughput. In an IoT & backend internship, I improved processing accuracy by 30% with enhanced validation and fault-tolerant error handling, and cut processing latency by 40% using event-driven refactoring.

On the GenAI side, I build inference systems that ship: I created CyberGuard, a production-grade AI moderation platform using fine-tuned MuRIL and Toxic-BERT (with Gemini API fallback), FastAPI inference microservices, and real-time moderation pipelines with Redis Pub/Sub and WebSockets. I also developed a Gemini-powered AI mock interview platform and built a production SSR e-commerce platform—always balancing model capabilities with robust APIs, async workflows, and clean deployment.