Open to opportunities

Ayush Gupta

@ayushgupta12

Message

Staff Software Engineer building high-scale AI-integrated real-time platforms.

India

Message

What I'm looking for

I’m looking to build and scale “Systems First” real-time products—observable, self-healing architectures with AI safety, RAG, and edge inference. I want ownership across backend, frontend, and infrastructure, with strong engineering governance.

I’m a Staff Software Engineer and Founding Engineer focused on the architecture of high-scale, AI-integrated real-time platforms. I’ve taken complex systems from 0-to-1, including multi-tenant Next.js applications supporting 500k+ monthly active users.

I build for real-time reliability and measurable performance—designing WebRTC state machines for hot-swapping hardware mid-call, and migrating inference to a Triton Inference Server cluster with Dynamic Batching to reach 1280+ RPS. I pair this with self-healing, observable systems using OpenTelemetry tracing and resilient backend workflows.

I also engineer AI safety and product-critical intelligence end-to-end: integrating LLM APIs for moderation and semantic search, and deploying a dual-track moderation pipeline with edge-local ONNX Runtime intelligence. Alongside that, I’ve architected Solana payment infrastructure and led teams through architectural governance, technical mentorship, and shipping AI-powered mobile experiences.

Experience

Work history, roles, and key accomplishments

Current

Staff Software Engineer

Current

Vooz Inc

Sep 2024 - Present (1 year 11 months)

Architected a multi-tenant Next.js environment and built the Vooz Video Coordinator state machine for reliable WebRTC handshakes and media recovery. Led Triton inference scaling, Solana payment infrastructure, LLM integration/orchestration, and multi-cloud GitOps/observability for high-throughput real-time experiences.

Next.js TypeScript WebRTC WebRTC State Machines Triton Inference Server Prometheus Grafana Solana SignalR Socket.IO ONNX Runtime WebGPU WASM LLM APIs Anthropic OpenAI RAG Prompt Engineering

Software Developer

Aimica Ltd

Jan 2023 - Sep 2024 (1 year 8 months)

Led a team of developers to build AI-powered mobile applications using React Native (Expo) and set architectural and code quality standards. Designed and implemented a scalable RAG-based knowledge system and backend services for AI orchestration, caching, and semantic retrieval.

React Native Expo Team Leadership RAG LLM APIs Anthropic: Claude Python Caching AI Orchestration Swift Java

Mobile Developer

Infoedge Software Solutions

Dec 2021 - Dec 2022 (1 year)

Optimized React Native flows for high-traffic applications by resolving memory leaks and improving UI frame rates on low-end devices. Collaborated on strictly typed REST API contracts to ensure consistency between mobile and server layers.

React Native Performance Optimization UI Frame Rate Optimization