Backend engineer specializing in high-performance systems, media pipelines, and observability at scale. I build reliable infrastructure that powers production systems handling millions of transactions, with a proven track record of reducing operational costs and shipping features with zero-downtime deployments.

Currently: Building video ingestion and processing pipelines at Skidos Previously: SDE-1 Backend at Samespace, working on AI-driven search and LLM infrastructure

• Video & Media Pipelines – HLS streaming, adaptive bitrate transcoding, CDN optimization
• Distributed Systems – Microservices architecture, async job orchestration, inter-service communication
• Observability & Monitoring – OpenTelemetry, SigNoz, Prometheus/Grafana, production alerting
• Infrastructure Optimization – Cost reduction through smart caching, data migration, indexing strategies
• AI/LLM Integration – Vector databases, semantic search, document-grounded conversations, multi-language localization

• 80% latency reduction – Edge caching strategy for notification delivery system (P99 latency)
• CDN cost optimization – On-demand segment delivery vs. full file transfers in HLS pipeline
• Significant DB cost savings – Cold storage migration + indexing overhaul for stale data management
• Days to minutes – Automated multi-language voice-over pipeline (14 languages via GPT-4 + ElevenLabs)
• 150+ APIs refactored – Domain-Driven Design implementation without production impact

• Email: sinhaditya50@gmail.com
• Portfolio: sinhaditya.com
• Resume: View on Google Drive

Building the backbone of a kids' educational platform with focus on video delivery and AI-powered content.

Key Projects:

• Video Ingestion Pipeline – Engineered end-to-end HLS adaptive bitrate streaming system, delivering chunks on-demand to reduce CDN bandwidth
• Multi-Language Voice-Over Pipeline – Automated content localization across 14 languages using GPT-4 rule engine, ElevenLabs synthesis, and CDN delivery (days to minutes turnaround)
• Cost Optimization – Migrated stale data to cold storage and overhauled indexing strategy, achieving significant managed DB cost reduction
• API Refactoring – Led phased migration of 150+ APIs to Domain-Driven Design (DDD) service layers with zero production impact

Developed AI-powered communication and search platform, handling real-time data at scale.

Key Projects:

• Notification Delivery System – High-throughput system with edge caching, achieving 80% P99 latency reduction and improved reliability at scale
• AI-Driven Search – Implemented semantic full-text and image retrieval using QdrantDB with document captioning and chunking
• Document-Grounded LLM – Built backend infrastructure with gRPC communication, vector retrieval (cosine similarity + confidence thresholding), and GPT-3 integration
• Production Observability – Implemented comprehensive monitoring with SigNoz and OpenTelemetry for real-time insights and alerting

Go, Docker, RabbitMQ, MinIO, PostgreSQL, Redis, Nginx, FFmpeg, Prometheus, Grafana

Scalable microservices-based video transcoding and delivery system.

• M3U8 chunked streaming with adaptive bitrate (480p/720p/1080p) and seamless quality switching
• RabbitMQ-driven async job orchestration with horizontally scalable FFmpeg workers
• MinIO (S3-compatible) for durable artifact storage
• Production-grade observability with Prometheus metrics and Grafana dashboards

Go, QdrantDB, gRPC, OpenTelemetry

AI-powered search engine for contacts and conversations.

• Document chunking and embedding generation
• Cosine similarity search with confidence thresholding
• Low operational overhead, self-serve infrastructure
• Full-text and image retrieval capabilities

B.Tech – Information Technology Kalinga Institute of Industrial Technology (KIIT), Bhubaneswar, Odisha Graduated 2024 | CGPA: 8.9

• Rust for systems programming and high-performance backend systems
• ELK Stack for enterprise-scale distributed logging
• Advanced Kubernetes and container orchestration patterns
• Rust async/await patterns and async runtime optimization

Passionate about sharing knowledge on distributed systems, scalability patterns, and production-grade engineering practices.

Let's collaborate on impactful backend systems! Feel free to reach out for discussions on distributed systems, system design, or potential opportunities. 🚀