Roshan Singha

Frontend Engineer

Gurugram, Haryana, India5 yrs 2 mos experience

Most Likely To SwitchAI Enabled

Key Highlights

Expert in building real-time, data-intensive UIs.
Led AI integration for enterprise digital twin platform.
Architected systems serving hundreds of kitchens.

Stackforce AI infers this person is a Frontend Engineer specializing in AI-driven applications and real-time systems.

Contact

Skills

Core Skills

ReactRag / Llm IntegrationAngular

Other Skills

RAG / LLM Integration (OpenAI, Anthropic, Google)React / Next.js (App Router, Server Components)SSE streamingvectorizationembedding storageJWTWebSocketdata visualizationReact QueryPusher WebSocketsCSV ingestionReact WindowHTML5Angular MaterialTypeScript

About

Frontend engineer specializing in real-time, data-intensive UIs — streaming AI interfaces, WebSocket-driven dashboards, and complex state management at scale. Currently building the frontend and AI integration layer for an enterprise digital twin platform (SSE streaming across OpenAI/Claude/Gemini, RAG-powered chat, credit metering). Previously: real-time kitchen operations system at Rebel Foods serving 400+ kitchens.

Experience

5 yrs 2 mos

Total Experience

1 yr 8 mos

Average Tenure

2 yrs 11 mos

Current Experience

Primathon

SDE-2

Jun 2023 – Present · 2 yrs 11 mos · Gurugram · Hybrid

AI & RAG
→ Designed RAG pipeline end-to-end — chunking, vectorization (40-concurrency queue), embedding storage; prompt-resolver augments LLM calls with retrieved context across OpenAI, Claude, and Gemini
→ Multi-provider LLM orchestration with SSE token streaming, provider-specific prompt formatting, and fallback routing on provider failures
→ Credit-gated AI metering — atomic pre-deduction with auto-refund on failure, handling race conditions across async provider calls
Frontend
→ Built the real-time AI chat interface with SSE streaming, inline data visualization, JWT refresh mutex, cross-tab session sync, and optimistic updates
→ Voice onboarding via SSE — 8+ phase state machine, event deduplication, recoverable errors, MediaRecorder capture with WebM→WAV conversion
→ Chat orchestration hook managing SDK lifecycle, WebSocket connections, JWT refresh, multi-session history, media streaming, citation extraction, and tool call routing
→ Dataset upload hook with exponential backoff reconnection, JWT expiry-aware refresh, batch completion detection — zero dropped uploads in production
→ WCAG 2.1 accessible modals with inert-based focus trapping, aria-live regions for streaming updates, and full keyboard navigation across all flows
→ Responsive dual-rendering via PlatformRouter — mobile/desktop component trees across 30+ routes with code-split bundles
Backend
→ SSE + WebSocket hybrid real-time layer scaled via Redis pub/sub across multiple container instances
→ SSE proxy bridging Python AI service to frontend — multi-subscriber fanout, event buffering with replay, heartbeat keepalive, and Redis-backed session recovery
→ 10 BullMQ processors for vector uploads (40 concurrency), conversation storage (50), KG extraction, and voice training — 92-table PostgreSQL schema
Hard Tradeoff
→ Fixed critical SSE bug where React StrictMode double-unmounts killed active streams — lifecycle refs distinguishing cleanup from teardown, eliminating false terminations

RAG / LLM Integration (OpenAI, Anthropic, Google)React / Next.js (App Router, Server Components)ReactRAG / LLM Integration

Rebel foods (formerly faasos)

SDE-1

Jul 2021 – Apr 2023 · 1 yr 9 mos · Bengaluru · Hybrid

Architected a real-time Kitchen Display & Order Management System (React, React Query, Pusher WebSockets) serving 400+ kitchens processing 10K+ daily orders — replaced ~120s polling with WebSocket push, reducing order-to-screen latency from ~120s to <1s and improving kitchen throughput during peak hours.
Built a polygon-based delivery zone editor (Google Maps API, geospatial polygon CRUD) used by ops teams across 70+ cities for 45+ brands — eliminated engineering dependency for zone changes, reducing turnaround from days (Jira → dev → deploy) to minutes via self-serve tooling.
Developed the product catalog & menu management system handling thousands of SKUs across 45+ brands with bulk CSV ingestion (5,000+ records/upload) and virtual scrolling (React Window) for 1,000+ row datasets — became the single source of truth for menu data powering all customer-facing apps.
Designed and shipped a cross-application RBAC system spanning 3 frontend apps and 50+ protected routes with role-based guards across 3 user tiers — enabled ops teams to onboard new brands without engineering tickets, supporting expansion from ~30 to 45+ brands.
Built and maintained a shared component library (80+ components, adopted by 4 teams org-wide) — standardized UI patterns across the platform, reducing new feature scaffolding time by ~30%.
Set up Datadog RUM dashboards tracking Core Web Vitals, API error rates, and page load times across 3 internal apps — reduced mean time to detect frontend regressions.

Reliance industries limited

Software Engineer

Aug 2019 – Feb 2020 · 6 mos · Mumbai Area, India

Frontend engineer on the JioMart e-commerce platform during its pre-launch build phase.
→ Built modular Angular components and reactive state flows (RxJS) for cart, checkout, and product catalog — part of the frontend foundation for a platform that scaled to 100M+ users post-launch
→ Integrated RESTful APIs with error boundaries and data normalization for catalogs with thousands of SKUs
Tech: Angular, TypeScript, RxJS, SASS, RESTful APIs, Git

HTML5Angular MaterialAngular