Gurugram, India · Open to impactful roles

I am Abdul Rehan

Senior Backend & AI Engineer

7+ years building production systems at xAI, Expedia, Razorpay, and Turing. I design high-throughput backends, fintech platforms, and agentic AI systems — Kafka at 5K+ msg/sec, LLM evaluation at 100K+ outputs/day, and products that drove ₹8 Cr GMV.

Get in touch GitHub LinkedIn Medium

About

Backend systems, AI agents, production scale

I'm a Senior Backend & AI Engineer focused on systems that stay reliable under real load — microservices, event-driven pipelines, payments, and LLM platforms. Recently at xAI I evaluate AI code-reasoning quality and build internal tooling that improves training feedback loops.

Before that I led high-throughput messaging and checkout modernization for Expedia (via EPAM), launched UPI-on-PPI and white-label wallets at Razorpay, and as an AI engineer at Turing shipped MCP servers, enterprise RAG, and LLM-as-a-Judge at production scale.

Education: B.Tech in Information Technology, UIET Kanpur (2015–2019).

Systems

Kafka, Spring Boot, Go services, Kubernetes, CI/CD, observability with Datadog & Splunk.

AI platforms

Agentic workflows, MCP integrations, RAG, vector DBs, LLM evaluation & grounding.

Fintech

UPI, wallets, payments auth, multi-gateway checkout, enterprise merchant platforms.

Experience

Where I've shipped impact

  1. Feb 2026 – Present Remote

    Software Engineer, Human Data · xAI

    • Evaluate AI code reasoning quality across languages and domains — design rubrics and structured feedback that improve model performance.
    • Built internal evaluation dashboard (React, FastAPI) and automated assessment workflows for data-driven training improvements.
  2. Mar 2024 – Jan 2026 Gurugram

    Software Engineer L2 · EPAM (Expedia as SDE 3)

    • Architected high-throughput messaging at 5K+ messages/sec with Apache Kafka and Java — +30% throughput, −15% errors.
    • Led monolith-to-microservices migration for Expedia checkout (Spring Boot, Docker) — 20% lower latency, 35% faster deploys.
    • Integrated global payment gateways (Apple Pay, Adyen, Naver Pay) and a centralized rules engine with Datadog/Splunk — +18% checkout success, −40% debug time.
  3. May 2024 – Jan 2026 Remote · Consulting

    AI Engineer · Turing

    • Built RL Gym platform with 30+ real-world evaluation workflows for autonomous e-commerce agents (Docker sandboxes, automated validation).
    • Designed MCP servers integrating Jira, Confluence, GitHub, Slack, Salesforce, Shopify, HubSpot, Square, SAP, and Zendesk.
    • Shipped enterprise RAG platform and LLM-as-a-Judge evaluating 100K+ outputs/day — 30% fewer hallucinations via retrieval and knowledge-graph grounding.
  4. Jan 2022 – Jan 2024 Bangalore

    Senior Backend Engineer · Razorpay

    • Launched flagship UPI-on-PPI (Java, Spring Boot, microservices) — ₹8 Cr GMV; enterprise clients including Swiggy and Indigo.
    • Built white-label wallet platform for B2B clients (Go, PostgreSQL) — +30% merchant adoption, 60% faster time-to-market.
    • Architected scalable auth with AWS Cognito and JWT — 10K+ concurrent users, 99.99% uptime, 40% fewer auth incidents via CI/CD.
  5. Jul 2020 – Jan 2022 Pune

    Backend Engineer · Smit.fit

    • Developed microservices health platform (Java, Spring Boot, MongoDB) for chronic disease management — scaled to 5K+ active users.
    • Engineered Razorpay payment integration (REST, OAuth2) — +20% conversion, 99.5% payment success.
    • Automated onboarding and analytics with Docker, Kubernetes, and CI/CD — 80% faster acquisition, +45% engagement.

Core competencies

Languages, cloud, AI, and reliability

Languages & frameworks

Java Python Go TypeScript JavaScript React Spring Boot FastAPI SQL

Cloud & infrastructure

AWS Docker Kubernetes Jenkins Terraform CI/CD Kafka Microservices

AI, ML & data

LLMs Agentic AI MCP RAG Vector DBs LangChain PostgreSQL MongoDB Redis

Security & monitoring

OAuth2 JWT Spring Security Datadog Splunk Prometheus Grafana

Selected work

Production impact & open projects

Highlights from industry work plus public repositories on GitHub.

xAI · AI evaluation

AI code-reasoning evaluation

Rubrics, structured feedback, and a React + FastAPI dashboard to streamline model assessment and training improvements.

React FastAPI LLMs Eval
Expedia · Scale

Checkout messaging & payments

Kafka at 5K+ msg/sec, microservices migration, and multi-gateway payments (Apple Pay, Adyen, Naver Pay) with full observability.

Java Kafka Spring Boot Datadog
Razorpay · Fintech

UPI-on-PPI & white-label wallets

Flagship prepaid UPI product (₹8 Cr GMV) and B2B white-label wallets in Go — enterprise onboarding for Swiggy, Indigo, and more.

Java Go PostgreSQL AWS Cognito
Turing · Agentic AI

MCP servers, RAG & LLM-as-Judge

30+ RL Gym workflows, enterprise MCP integrations, and production RAG evaluating 100K+ outputs/day with 30% fewer hallucinations.

MCP RAG Docker Python
Open source

Screen-It

Video calls, screen sharing, and live chat with Java Spring Boot and WebRTC.

View on GitHub →
Spring Boot WebRTC JavaScript
Open source

Khata Book

Personal finance / ledger-style Java application for tracking transactions and balances.

View on GitHub →
Java Backend
Open source

Umee App

Java-based application work exploring service design and backend patterns.

View on GitHub →
Java
Open source

More on GitHub

Experiments, utilities, and contributions — weather apps, CLI tools, and infrastructure forks for learning production systems.

Browse repositories →

Contact

Let's build something solid

Whether it's high-scale backends, payments infrastructure, or agentic AI platforms — I'd love to hear about the problem you're solving.

Sends directly to my inbox. Prefer email? Write to uiet.rehan@gmail.com.