Bigya Tuladhar
I’m a Senior Full-Stack & Cloud Engineer specializing in production Generative AI on AWS—RAG, agentic workflows, and reliability engineering for real-time platforms at scale.
About
I’m a Senior Full-Stack & Cloud Engineer specializing in production Generative AI on AWS—RAG, agentic workflows, and reliability engineering for real-time platforms at scale. I combine deep AWS serverless/IaC experience with full-stack delivery to build responsive, source-grounded AI features.
I’ve led end-to-end work from architecture and telemetry to low-latency streaming and operational robustness, including cross-region inference failover, guardrailed agent systems, and scalable real-time experiences serving thousands of concurrent users.
- 8 years' experience
- Joined August 2026
Skills· 16
Languages
- EnglishAdvanced
- NepaliFluent
Industry experience
- Media & Entertainment
- Software & Internet
- Computers & Electronics
More experts in Kathmandu, Nepal
Work experience· 8 roles
Integrated third-party game provider APIs in Node.js (session management, webhook callbacks, real-time balance updates). Built a secure Stripe wallet system for deposits, payouts, and reconciliation.
Integrated third-party game provider APIs with Node.js for session management, webhook callbacks, and real-time balance updates. Built a secure Stripe wallet system for deposits, payouts, and reconciliation.
Built a production RAG platform on AWS Bedrock powering a live moderator copilot plus live/post-event AI assistants and livestream summarization across 50+ events, with grounded answers and source-backed accuracy. Engineered cross-region Bedrock inference failover with exponential backoff and token quota handling to reduce throttling failures by ~68% under burst load. Designed an agentic workflow for event-metrics natural-language queries using LLM-generated Athena/DynamoDB queries, validator-driven refine-retry loops, and guardrails via read-only IAM and tenant-scoped filters, with deterministic fallbacks. Re-architected real-time interactions using AWS AppSync (GraphQL subscriptions) and DynamoDB, migrating off AWS IoT Core to support production scale with thousands of concurrent viewers. Delivered ultra-low-latency streaming via Amazon IVS over WebRTC with real-time transcription/translation, HLS transcoding via MediaConvert, and KMS encryption at rest; built the Mission Control das
Education
Hire Bigya Tuladhar today
Post a job and invite Bigya, or contact them directly. No hidden commission.







