Open to opportunities

Harsh Kedia

@harshkedia1

Message

Senior product engineer building AI-native platforms and high-performance backend systems end to end.

India

Message

What I'm looking for

I’m looking to own end-to-end AI product engineering—shipping reliable, production-grade systems with fast iteration, strong experimentation, and measurable performance improvements across backend, AI, and user experience.

I’m a Senior Product Engineer who builds AI-native products and the backend systems that make them reliable, fast, and scalable. At Shoppin, I shipped an AI-native custom-clothing platform end to end—conversational requirement gathering, live design generation, automatic pricing, and production handoff.

I’ve also delivered high-impact AI features in production: an agent-based pattern-making engine that generates production-ready ASTM sewing patterns at ~95% accuracy in minutes, and an AI virtual try-on experience for 20K+ daily users (14x faster: 6.8s to 500ms). I owned the recommendation & discovery engine powering a personalized feed of 250K+ items, and built the entire e-commerce & payments stack, including order lifecycle, integrations (Razorpay, PayPal, Stripe), and advanced commerce flows.

Previously, as a Founding AI Engineer at Verbaflo.AI, I built an agentic multi-modal AI backend with FastAPI (scaling to 20,000+ concurrent sessions), architected a hybrid RAG pipeline with Milvus (sub-200ms retrieval, 95%+ context accuracy), and shipped production LLM infrastructure with multi-provider support via gRPC. Earlier, at PrudentBit, I architected a secure file-sharing platform with enterprise security (SAML SSO, OAuth, 2FA, DLP, malware scanning) and rebuilt legacy services for major latency improvements.

Experience

Work history, roles, and key accomplishments

Current

Senior Product Engineer

Current

Shoppin

Oct 2025 - Present (9 months)

Built an agent-based pattern-making engine that converts garment images/specs into production-ready ASTM sewing patterns, and shipped an AI-native custom-clothing platform end to end (requirements to design generation to production handoff). Owned the e-commerce/payments stack, recommendation/discovery engine, and led AI virtual try-on for daily users.

AI Agents Pattern Making Design In Browser Payments Razorpay PayPal Stripe Order Tracking Coupons Cross Gateway Refunds and Cancellations Personalized Video and Image Feed

AI Engineer

Sidecar

Jul 2025 - Oct 2025 (3 months)

Built a 24/7 autonomous freight-forwarding agent that reads inbound emails, uses an LLM for intent understanding and structured extraction from messy documents, and auto-responds to collect missing shipment details with long-term memory of client preferences. Designed the agent harness and an agentic quoting loop that consults partner freight-forwarder agents, ranks options, and returns a best mat

Email Ingestion Structured Extraction Long Term Memory Layer Agent orchestration Tool Use Retries GuardRails Autonomous Agents Ranking Price Reliability

Founding AI Engineer

Verbaflo.AI

Dec 2024 - Sep 2025 (9 months)

Built the backend for a multi-modal agentic AI platform (Chat, Email, WhatsApp, Voice) including real-time messaging, scaling to 20,000+ concurrent sessions. Architected a hybrid RAG pipeline with Milvus for low-latency retrieval, and shipped production LLM infrastructure with multi-provider support via gRPC.

fastAPI Milvus Hybrid RAG Reranking gRPC OpenAI Anthropic LLaMA Qwen Evaluation Pipelines

Founding Engineer

PrudentBit

Jun 2023 - Aug 2024 (1 year 2 months)

Architected a secure file-sharing platform on Azure, deploying multiple microservices with enterprise security controls (SAML SSO, OAuth, 2FA, DLP, and malware scanning) for Toyota and UCO Bank. Rebuilt and scaled a legacy backend with REST APIs, Celery/RabbitMQ queues, chunked uploads, full-text search, and a RAG-powered document interface.