Deipayan Dash profile photo 1 on Tal

Deipayan Dash

Engineering @ Oracle | Ex-Razorpay, JioCinema | Distributed Systems

Bengaluru

5+ years of work experience

Oracle
Aug 20251 yr 2 mosPresent

Bengaluru, Karnataka, India

• Built OCI AI Vision’s real-time video inference platform for object detection and face identification across global regions. • Migrated video decoding and inference components from Python to Go and C++ with NVIDIA DeepStream. • Reached sub-30 ms p90 latency and about 70% higher throughput on NVIDIA A10 GPU clusters. • Worked across RTSP/WebRTC streaming, Kubernetes, gRPC, GPU utilization and observability.
Razorpay
Jul 20241 yr 1 moJul 2025

Bengaluru, Karnataka, India

• Re-architected core UPI payment services for modularity, reliability and feature scalability. • Cut latency by 90% on failed-payment gateway flows. • Improved auto-reconciliation across multiple gateways and strengthened security with hybrid encryption. • Built Prometheus and Grafana monitoring for gateway performance and stability.
JioHotstar
Nov 20239 mosJul 2024

Bengaluru, Karnataka, India

• Built an HLS and DASH stream-analysis system for JioCinema’s live-streaming platform. • Supported 15M+ concurrent viewers during IPL 2024. • Used FFmpeg to validate segments and detect faulty playlists, with real-time monitoring to prevent playback failures during major live events. • Monitored SSAI streams across assets, languages and platforms using Prometheus and Grafana.
Capita
Mar 20212 yrs 9 mosNov 2023

Pune Division, Maharashtra, India

• Built internal web services and REST APIs used by 50K+ daily users across five countries. • Developed backend services with Spring Boot and Express and product features in Angular. • Improved performance, responsiveness, caching, accessibility and modularity. • Took ownership of a key product and was promoted to Feature Owner.
Real Dev Squad
Dec 20202 yrs 11 mosOct 2023

Remote

Deipayan’s AI Stack

claude code
claude code
codex
codex

Education

Maulana Abul Kalam Azad University of Technology, West Bengal formerly WBUT

Bachelor of Technology, Electrical, Electronics and Communications Engineering

Maulana Abul Kalam Azad University of Technology, West Bengal formerly WBUT

20164 yrs2020

Join Deipayan and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Deipayan Dash on Tal

Headline

Engineering @ Oracle | Ex-Razorpay, JioCinema | Distributed Systems

About

Backend and distributed systems engineer with 5+ years of experience building low-latency, high-scale platforms. At OCI AI Vision, I build real-time video inference systems in Go, C++ and Python using NVIDIA DeepStream, Kubernetes, gRPC and GPU infrastructure. I migrated core video decoding and inference components from Python to Go/C++, reaching sub-30 ms p90 latency and about 70% higher throughput on GPU clusters. Previously, I re-architected Razorpay's core UPI payments service and cut latency by 90% on failed-payment gateway flows. At JioCinema, I worked on live-streaming infrastructure supporting 15M+ concurrent viewers during IPL 2024. At Capita, I built internal services used by 50K+ daily users across five countries and was promoted to Feature Owner. I focus on backend systems, distributed systems, performance, reliability, observability and AI/GPU inference.

Location

Bengaluru

AI stack

claude code

codex

Work experience

Software Engineer III-AI Vision at Oracle, Aug 2025 - Present · 1 yr 2 mos, Bengaluru, Karnataka, India, • Built OCI AI Vision’s real-time video inference platform for object detection and face identification across global regions. • Migrated video decoding and inference components from Python to Go and C++ with NVIDIA DeepStream. • Reached sub-30 ms p90 latency and about 70% higher throughput on NVIDIA A10 GPU clusters. • Worked across RTSP/WebRTC streaming, Kubernetes, gRPC, GPU utilization and observability.

Software Engineer at Razorpay, Jul 2024 - Jul 2025 · 1 yr 1 mo, Bengaluru, Karnataka, India, • Re-architected core UPI payment services for modularity, reliability and feature scalability. • Cut latency by 90% on failed-payment gateway flows. • Improved auto-reconciliation across multiple gateways and strengthened security with hybrid encryption. • Built Prometheus and Grafana monitoring for gateway performance and stability.

Software Development Engineer 2-Jio Cinema at JioHotstar, Nov 2023 - Jul 2024 · 9 mos, Bengaluru, Karnataka, India, • Built an HLS and DASH stream-analysis system for JioCinema’s live-streaming platform. • Supported 15M+ concurrent viewers during IPL 2024. • Used FFmpeg to validate segments and detect faulty playlists, with real-time monitoring to prevent playback failures during major live events. • Monitored SSAI streams across assets, languages and platforms using Prometheus and Grafana.

Software Developer at Capita, Mar 2021 - Nov 2023 · 2 yrs 9 mos, Pune Division, Maharashtra, India, • Built internal web services and REST APIs used by 50K+ daily users across five countries. • Developed backend services with Spring Boot and Express and product features in Angular. • Improved performance, responsiveness, caching, accessibility and modularity. • Took ownership of a key product and was promoted to Feature Owner.

Open Source Contributor at Real Dev Squad, Dec 2020 - Oct 2023 · 2 yrs 11 mos, Remote

Education

Bachelor of Technology, Electrical, Electronics and Communications Engineering, Maulana Abul Kalam Azad University of Technology, West Bengal formerly WBUT