r
rohittcodes

Rohith Singh

@rohittcodes

AI Agent Interface and Workflow Engineer, Full Stack TypeScript Specialist

Índia
Inglês, Hindi
Algumas informações são exibidas no idioma inglês.
Sobre mim
Stop building basic AI wrappers. I architect production-grade AI Agent Interfaces and Workflow Builders that handle complex logic, streaming data, and multi-agent coordination. I solve the friction between LLM backends and user experience. From AI-native ledgers like Linea to drag-and-drop builders, I focus on performance and technical clarity. • Performance: Optimized React apps from 43 to 100 Lighthouse. • Scale: Architected AI/Web3 platforms serving 5k+ users. • Tech: TypeScript, Node.js, and Agentic RAG Pipelines. Let’s discuss your architecture. no fluff, just delivery.... Saiba mais

Habilidades

r
rohittcodes
Rohith Singh
offline • 

Portfólio

Experiência profissional

Product Engineer | AI Architect Engineer

createxp • Período integral

Oct 2025 - Present • 7 mos

• Designed and deployed a production RAG pipeline using FastAPI, PostgreSQL/pgvector, Redis, and Sentence Transformers (intfloat/e5-base-v2) for semantic document retrieval and AI-powered recommendations. • Built LLM-based intent extraction layer using Gemini Flash 3 to convert natural language queries into structured retrieval context, enabling streaming API responses with sub-200ms latency. • Implemented multi-signal ranking system combining vector similarity, spec matching, and contextual scoring, packaged as Dockerized microservice with webhook support, authentication, and structured logging.

Software Development Engineer

Grit Labs • Período integral

Jul 2025 - Oct 2025 • 3 mos

• Architected and deployed fam.cool, a multi-agent Web3 chat platform serving 5K+ users with real-time messaging, AI-powered conversations, and blockchain integration using Next.js, PostgreSQL, Redis, and RabbitMQ • Migrated legacy React application to Next.js improving Core Web Vitals from 43 to 100, reducing bundle size by 45% and implementing Docker containerization for DigitalOcean deployment with tarball optimization

Full Stack Developer

createxp • Período integral

May 2025 - Jul 2025 • 2 mos

• Engineered high-performance CRM APIs processing 15K+ requests/day using Node.js, TypeScript, and PostgreSQL with 99.5% uptime and sub-100ms response times • Implemented vector search capabilities with pgvector enabling semantic document retrieval and AI-powered client matching, increasing conversion rates by 25%