Rajendra is available for hire
Hire RajendraRajendra Upadhyay
Verified Expert in Engineering
Full-stack Developer
New Delhi, Delhi, India
Toptal member since February 23, 2022
Bio
Rajendra is a full-stack and AI engineer with a decade of experience in React, React Native, and Node.js, now building and leading production-agentic AI systems across the OpenAI and Anthropic stacks. He architects RAG and multi-step agent workflows, owns LLM evaluation and cost, and mentors engineers, shipping polished React interfaces and reliable, observable agents with equal fluency in TypeScript and Python.
Portfolio
Clevertap
TypeScript, Node.js, Python, Anthropic SDK, Model Context Protocol (MCP)...
Dust
TypeScript, React, Python, FastAPI, Anthropic, Model Context Protocol (MCP)...
B2BBricks
TypeScript, Next.js, React, Node.js, PostgreSQL, Prisma, Tailwind UI, Cursor AI...
Experience
- React - 9 years
- TypeScript - 9 years
- JavaScript - 9 years
- Redux - 8 years
- Material UI - 6 years
- CSS - 6 years
- HTML - 6 years
- React Native - 6 years
Preferred Environment
MacOS, Linux, Windows
The most amazing...
...project I've built is a production agent platform — RAG, tool-calling, MCP servers, and evals, that automates real workflows end to end.
Work Experience
Lead AI Engineer
2025 - PRESENT
Clevertap
- Led AI engineering for a data-enrichment platform, setting agent architecture standards and mentoring a team of four engineers.
- Architected an agentic RAG system that matches and enriches entities across noisy third-party datasets using LLM reasoning over a pgvector store in PostgreSQL.
- Built custom MCP (Model Context Protocol) servers to expose internal tools and data to Claude-based agents in a governed, reusable way.
- Designed autonomous multi-step agent workflows with the Anthropic SDK, combining tool-calling, retries, and human-in-the-loop checkpoints for high-stakes data decisions.
- Established an LLM evaluation and observability pipeline with golden sets, regression gates, and per-run cost and latency tracking.
- Introduced model routing and prompt caching across Claude and AWS Bedrock to cut inference spend while holding response quality steady.
- Drove company-wide adoption of Claude Code and CI agent harnesses, raising engineering delivery velocity.
- Built Python, Node.js, and TypeScript services to run agents reliably in production with structured logging and graceful degradation.
- Defined reusable agent patterns and internal guardrail libraries adopted across teams to standardize how agents are built and reviewed.
Technologies: TypeScript, Node.js, Python, Anthropic SDK, Model Context Protocol (MCP), Agentic RAG Systems, Pgvector, Claude Code, PostgreSQL, Cursor AI, Google Antigravity, OpenRouter, GraphQL, APIs, Front-end Development, Front-end, CSS Grid, Docker, Amazon Web Services (AWS)
AI Engineer (IC)
2024 - 2025
Dust
- Built multi-step AI agents on the Dust platform that automate internal knowledge retrieval and workflow tasks across connected data sources like Slack, Notion, and Google Drive.
- Implemented retrieval-augmented (RAG) pipelines over a pgvector store in PostgreSQL, tuning chunking and re-ranking to ground agent responses in company documents and reduce hallucinations.
- Developed tool-calling integrations with the Anthropic SDK in TypeScript that let agents query internal APIs, databases, and third-party services within scoped permissions.
- Built MCP (model context protocol) servers to expose internal tools and data sources to agents in a reusable, governed way.
- Designed structured-output schemas and prompt-engineering guardrails to keep agent actions deterministic, auditable, and safe to run unattended.
- Built Python and FastAPI services on AWS to orchestrate long-running agent jobs with retries, idempotency, and human-in-the-loop review queues.
- Added streaming responses and an observability layer that traces token usage, latency, and cost per agent run to keep production spend predictable.
- Built React interfaces for non-technical users to configure, run, and monitor agents without writing code.
- Established an evaluation harness with golden datasets and regression gates to catch agent quality drops before release.
- Adopted Claude Code into the team's daily workflow to accelerate feature delivery and shorten PR cycle time.
Technologies: TypeScript, React, Python, FastAPI, Anthropic, Model Context Protocol (MCP), Cursor AI, Claude Code, Agentic AI, Agentic Coding, GraphQL, APIs, Front-end Development, Front-end, CSS Grid, Docker, Amazon Web Services (AWS)
AI Engineer (Full-stack)
2023 - 2024
B2BBricks
- Built an AI document-processing pipeline in Python that extracts, classifies, and validates real-estate contracts, cutting manual review time by roughly 70%.
- Integrated GPT-4 through the OpenAI API using function calling and JSON mode for structured, deterministic field extraction.
- Implemented a retrieval-augmented (RAG) layer with LangChain and pgvector over PostgreSQL to ground extraction in prior deal data and reduce hallucinated fields.
- Generated and indexed OpenAI embeddings to power semantic search across historical contracts.
- Developed a FastAPI service to orchestrate long-running extraction jobs with retries, idempotency, and human-in-the-loop review queues.
- Adopted GitHub Copilot and Cursor into the team's workflow to accelerate feature delivery and shorten PR cycle time.
- Built an evaluation harness in Python with golden datasets to catch extraction regressions before release.
- Integrated React and TypeScript review dashboards for operators to approve, correct, and re-run pipeline outputs.
Technologies: TypeScript, Next.js, React, Node.js, PostgreSQL, Prisma, Tailwind UI, Cursor AI, OpenAPI, Model Context Protocol (MCP), Playwright, Vercel, OpenAI GPT-4 API, LangChain, Pgvector, RAG Systems, AWS Copilot, MongoDB, APIs, Django, Front-end Development, Front-end, Docker, Amazon Web Services (AWS)
Full-stack Developer
2022 - 2023
Allegis - Cruise Automation - Main
- Developed an internal management system for automated car logging and tracking.
- Implemented a smooth animation of a car moving, turning, and halting on a 2D map using Lottie and the animation library.
- Migrated most of the React components to React Native to build an app version for the trainers of the automated cars.
- Worked as an individual contributor on the landing page and business-facing website with smooth animation effects.
- Built the customized dashboard for the internals to keep track of KPIs and add a gateway for manual intervention by an authorized person.
- Helped the team optimize the web app's performance and accelerated the project's development.
Technologies: React, Node.js, TypeScript, PostgreSQL, Material UI, 3D Animation, Lottie, Redux, Storybook, React Native, APIs, Django, Front-end, Docker, Amazon Web Services (AWS)
Senior Software Engineer
2019 - 2022
A Short Video App Company
- Collaborated with clients and business analysts to determine software needs and system implications. Conducted and managed product launches and post-launch support using conference calls to coordinate with geographically spread teams.
- Built a robust approval system at the admin dashboard level using React.
- Participated in issue troubleshooting with various teams to identify and resolve root causes.
- Integrated the payments and payouts feature in the app with user-friendly designs.
- Participated in code reviews, screening interviews, and mentoring of junior developers.
- Led a team of three developers to migrate the app from React to React Native for web compatibility.
- Implemented 3D animation with the help of Lottie and the React animation library for visualization.
- Built a native module in React Native web to handle custom navigation both in the app and web version.
- Built and deployed the app in the iOS App Store and Google Play Store.
Technologies: React, React Native, Redux, Redux-Saga, JavaScript, HTML, CSS, TypeScript, Agile, Scrum, Agile Sprints, Style Guides, Lottie, 3D Animation, React Native for Web, GraphQL, MongoDB, Django, Front-end, CSS Grid, Docker, Amazon Web Services (AWS)
Senior Software Engineer
2018 - 2019
BITCS (Contract)
- Developed and deployed solutions focused on end-user experience, accessibility, and high availability with low latency and scalability.
- Integrated REST APIs with front-end dashboards for real-time visibility of details for customers.
- Identified and removed transaction processing bottlenecks to enable increased website traffic.
- Built a testing setup for the dashboard using Mocha, Enzyme, and Jest.
- Designed and launched dashboards for customers to see and manage their bank accounts using React, styled-components, and Redux.
Technologies: React, Redux, React Native, Express.js, Node.js, Mocha, Jest, Enzyme, UI Testing, Redis, Django, Front-end Development, Front-end, CSS Grid
Software Engineer
2018 - 2019
BITCS (Contract)
- Worked on digitizing the physical copies of the records to build multiple tools used in the supply chain industry. The project included creating and maintaining various internal tools.
- Built a CI/CD pipeline to ensure the products were delivered on time.
- Developed tools for monitoring and alerting in-house application operations.
- Coded a new front-end functionality using React, Angular, and Node.js to prepare reusable styles.
- Wrote web pages and delivered cross-browser compatibility using React, TypeScript, and React Native.
- Built a custom component library in React Native for the web that can be used in React Native as well as React.
- Helped in writing reusable and customizable React Native components that can be used in the browser as well, ensuring stricter code compatibility and less time.
Technologies: React, Node.js, HTML, JavaScript, TypeScript, Material UI, Redux, TypeScript 3
Software Engineer
2017 - 2018
BITCS (Contract)
- Built a highly scalable manufacturing-related solution. Developed a specialized eCommerce platform for gold company owners to acquire gold and silver ornaments straight from the producer.
- Implemented the structure and design of web app sections.
- Converted wireframes and mockups into fully functioning mobile-first web applications.
- Developed reusable front-end styled-components to enhance the user interface.
- Worked on the performance optimization of the web app.
Technologies: HTML, CSS, JavaScript, TypeScript, React, React Native, UI Style Guides, Material UI, Firebase, REST, Front-end Development
Software Engineer
2016 - 2017
A SaaS Product Company
- Built front-end features and contributed consistently to all parts of the codebase.
- Participated in scrum calls and followed agile methodologies in delivering user-centric products using React, Angular, Node.js, and React Native.
- Collaborated closely with product and customer success teams to understand pain points and develop end-to-end solutions.
- Swiftly troubleshot and addressed critical bugs.
- Supported the app's infrastructure to maintain reliability and scalability.
Technologies: JavaScript, Redux, React, Node.js, React Native, HTML, UI Testing, Redis
Experience
InstaGYM
InstaGYM was a web application for gym owners and their enthusiastic fitness customers. It had blogging features that gym owners administered. Users could be added, edited, and deleted by the gym management.
Greenkart
I built an inventory management system to track order details, categorize products with vendor management, and include user access management dashboards. It also had features like inventory alerts for maintaining operational control.
Education
2013 - 2017
Bachelor's Degree in Computer Science
Chandigarh University - India
Skills
Libraries/APIs
React, Node.js, Redux-Saga, React Native for Web, OpenAPI, Playwright
Tools
Mocha, Prisma, AWS Copilot, Claude Code
Languages
HTML, CSS, JavaScript, TypeScript, TypeScript 3, Python, GraphQL
Frameworks
React Native, Redux, Angular, Material UI, Express.js, Jest, Next.js, Django
Paradigms
Agile, Scrum, REST, Model Context Protocol (MCP)
Platforms
MacOS, Linux, Windows, Firebase, Android, Vercel, Docker, Amazon Web Services (AWS), iOS
Storage
PostgreSQL, Cloud Firestore, Redis, MongoDB
Other
Agile Sprints, Style Guides, UI Style Guides, Enzyme, TypeORM, UI Testing, Lottie, Storybook, Tailwind UI, Cursor AI, OpenAI GPT-4 API, LangChain, Pgvector, RAG Systems, FastAPI, Anthropic, Agentic AI, Agentic Coding, Anthropic SDK, Agentic RAG Systems, Google Antigravity, OpenRouter, APIs, Front-end Development, Front-end, CSS Grid, 3D Animation
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring