# Hrishabh Jain

> Hrishabh Jain is an AI Engineer in Bengaluru, India, building AI agents, RAG pipelines, evals and observability for production AI at CambrianEdge.ai. 3.5+ years shipping production software.

Hey, I'm Hrishabh Jain, an AI engineer in Bengaluru who teaches LLMs to show up to work on time. At CambrianEdge.ai I own the multi-agent AI core: 10+ agents that plan, research and write together under a supervisor that's honestly stricter than most managers. I build RAG pipelines that cite their sources instead of making things up, MCP integrations that let agents use real tools, and the TypeScript / MERN backend plumbing that keeps it all fast. I once cut response latency by ~40% just by convincing the platform it didn't need the biggest model for everything. Before that I was at Gutenberg, turning piles of RFPs into something you could just ask questions to. Basically, I take AI ideas out of demo-land and into production. Off the clock, I've solved 250+ LeetCode problems and will happily argue that a smaller model would've done the job.

## Quick facts
- Role: AI Engineer at CambrianEdge.ai
- Location: Bengaluru, Karnataka, India
- Experience: 3.5+ years
- Focus: Multi-agent systems & RAG
- Email: hrishabh507@gmail.com
- Website: https://hrishabh.in

## Experience
### AI Engineer, CambrianEdge.ai (Jan 2025 – Present)
Full-time · Remote. Own the multi-agent AI core of a production AI marketing platform for autonomous content generation and workflow automation.
- Architected a multi-agent system on Mastra with 10+ agents in a supervisor pattern, handling task decomposition, research workflows and autonomous execution.
- Built MCP (Model Context Protocol) integrations and reusable agent skills so agents can securely call external tools, APIs and data sources with structured outputs.
- Designed advanced RAG pipelines with LlamaIndex and LlamaParse: multimodal, layout-aware ingestion, semantic retrieval and context injection that cut hallucinations.
- Benchmarked LLM selection across the platform, swapping oversized models for task-appropriate ones and cutting end-to-end latency by ~40%.
- Built the agent orchestration layer that manages prompt routing, memory, tool usage and execution order for multi-step reasoning.
- Engineered LLM execution pipelines tuned for latency, cost and reliability across streaming and batch generation, plus real-time token streaming for long-form content.
- Added prompt abstraction and versioning so the team can experiment fast while keeping production behaviour reproducible.
- Designed MongoDB schemas, indexes and aggregations for high-throughput AI workloads.
- Secured REST and GraphQL APIs with enterprise-grade authentication, authorization and abuse prevention.
- Used Redis caching and BullMQ queues for event-driven, fault-tolerant execution of long-running AI jobs.

### Full-Stack Developer, Gutenberg (May 2024 – Dec 2024)
Full-time · Remote. Built AI-powered document intelligence: semantic search and question answering over large sets of RFPs.
- Built a question-answering system over RFP documents using semantic search, so users could query large document sets with high relevance.
- Implemented ingestion, chunking, embedding and retrieval pipelines with LangChain, OpenAI APIs and Pinecone, laying the groundwork for RAG.
- Compared vector search strategies, similarity metrics and chunking approaches to improve retrieval quality and reduce hallucinations.
- Designed and tested AI proofs of concept to evaluate emerging LLM APIs, frameworks and tooling for marketing and content problems.
- Experimented extensively with prompt engineering and response structuring to map LLM strengths and failure modes.
- Contributed to Node.js, Express and MongoDB backend services for internal tools and experimental AI features.

### Full-Stack Developer Intern, Gutenberg (Feb 2023 – Apr 2024)
Internship · Remote. Built backend APIs and real-time, role-based dashboards for the finance team, and early OpenAI prototypes.
- Designed scalable REST APIs with Node.js, Express and MongoDB behind a responsive React frontend with real-time updates.
- Built role-based dashboards with real-time insights that helped the finance team make faster, more accurate decisions.
- Added secure authentication, audit logs and data validation for compliance and traceability.
- Explored OpenAI APIs from their initial public release, building proofs of concept for AI-driven automation inside internal tools.

### Web Developer Intern, Vocally (Oct 2022 – Jan 2023)
Internship · Vancouver, Canada · Remote. Shipped full-stack features across a React frontend and a Node.js backend.
- Developed the React front-end architecture behind new user-interface concepts.
- Integrated frontend components with Node.js, Express, MongoDB and Mongoose backends.
- Built an internal admin panel with Retool and MongoDB.

### Technical Content Engineer Intern, Kalvium (Jun 2022 – Aug 2022)
Internship · Coimbatore, India · Remote. Designed hands-on web development labs and industry-driven projects for outcome-based learning.
- Created labs and practice exercises in HTML, CSS, JavaScript and React.
- Built projects from industry inputs to support outcome-driven learning.

## Projects
- Widgetsy (2022): A platform of customizable widgets for React apps, with a visual editor that generates the code and an npm package to drop them in. Built with Next.js, Node.js, Express, MongoDB, Docusaurus.
- Doodlesy! (2022): A real-time multiplayer drawing game with three modes: Guess It, Rate It and Grand Reveal. Built with React, Socket.io, Firebase.

## Writing
All writing: https://hrishabh.in/writing/
- [Why AI Didn't Cite Your Site in Search Results](https://www.linkedin.com/posts/hrishabh-jain_geo-aisearch-jev-activity-7509229400853254144-JFAy) (LinkedIn, 2026-09-25)
- [asked AI for the best whey protein in india 🏋️](https://x.com/hrishabh_hj/status/2103469380017492453) (X, 2026-09-25)
- [Database replication is the process of copying data from one database to another. Replicat](https://x.com/hrishabh_hj/status/1771545757520138509) (X, 2024-03-23)
- [Webhooks enable seamless communication between software systems, automating processes base](https://x.com/hrishabh_hj/status/1771154318227394958) (X, 2024-03-22)
- [Elevate software efficiency with message queues! They enable seamless communication, async](https://x.com/hrishabh_hj/status/1770654062104510515) (X, 2024-03-21)
- [Explore the top load balancing algorithms - Round Robin, Least Connections, Weighted Round](https://x.com/hrishabh_hj/status/1770305115418739127) (X, 2024-03-20)
- [Redis turbocharges Node server caching, optimizing performance & scaling. When a request c](https://x.com/hrishabh_hj/status/1769929286709125539) (X, 2024-03-19)
- [Benefits of using a Reverse Proxy for optimized security, efficiency, and reliability in y](https://x.com/hrishabh_hj/status/1769566901011935455) (X, 2024-03-18)
- [Reverse proxy is like a superhero guarding your servers from evil internet baddies, while ](https://x.com/hrishabh_hj/status/1769263613653741590) (X, 2024-03-17)
- [Bringing Characters to Life: The Power of Generative Agents](https://medium.com/@HrishabhCodes/bringing-characters-to-life-the-power-of-generative-agents-in-ai-6a471737acd0) (Medium, 2023-04-18)
- [How to add Conversational UI in React Apps using Alan AI](https://medium.com/@HrishabhCodes/how-to-add-conversational-ui-in-react-apps-using-alan-ai-959b839cda42) (Medium, 2022-11-11)

## Skills
- Gen AI: Mastra, MCP, Agent Skills, Multi-Agent Systems, RAG, Context Engineering, Agentic Memory, Vercel AI SDK, LlamaIndex, LlamaParse, LangChain, OpenAI APIs
- Backend: Node.js, Express, REST APIs, Apollo GraphQL, WebSockets, BullMQ
- Frontend: React, Next.js, Redux, Zustand, Tailwind CSS
- Data & Cloud: MongoDB, Redis, Pinecone, Neo4j, Firebase, Docker, AWS, Azure
- Languages: TypeScript, JavaScript, Python, Java
- Fundamentals: DSA, System Design, DBMS, OOP, Operating Systems, Computer Networks, Git

## Education
- Bachelor of Technology in Computer Science, Amity University Rajasthan, Jaipur, India (2020–2024), CGPA 8.45
- Certification: Generative AI Leader (Google)
- Certification: Crash Course on Python (Google)
- Certification: Neo4j Fundamentals (Neo4j)
- Certification: Cypher Fundamentals (Neo4j)
- Certification: Introduction to DevOps

## FAQ
### Who is Hrishabh Jain?
Hrishabh Jain is an AI Engineer based in Bengaluru, India, with 3.5+ years of experience building production software. He currently works at CambrianEdge.ai, where he architects and owns the platform's multi-agent AI core.

### What does Hrishabh Jain work on?
He builds production-grade AI systems: multi-agent orchestration with 10+ agents in a supervisor pattern, retrieval-augmented generation (RAG) pipelines with LlamaIndex and LlamaParse, Model Context Protocol (MCP) integrations, and real-time token streaming. He also benchmarks LLM selection, which cut end-to-end latency by about 40%.

### What is Hrishabh Jain's tech stack?
TypeScript and the MERN stack (MongoDB, Express, React, Node.js) are his foundation, extended into AI with Mastra, MCP, Vercel AI SDK, LlamaIndex, LangChain, Pinecone and OpenAI APIs. For infrastructure he uses Redis, BullMQ, GraphQL, Docker, AWS and Azure.

### Where did Hrishabh Jain study?
He holds a B.Tech in Computer Science from Amity University Rajasthan, Jaipur (2020–2024), graduating with a CGPA of 8.45. He is also a Google Generative AI Leader certified professional.

### How can I contact Hrishabh Jain?
Email hrishabh507@gmail.com, or reach him on LinkedIn (linkedin.com/in/hrishabh-jain) or X (@hrishabh_hj).

## Links
- [GitHub](https://github.com/HrishabhCodes)
- [LinkedIn](https://www.linkedin.com/in/hrishabh-jain/)
- [X](https://x.com/hrishabh_hj)
- [LeetCode](https://leetcode.com/u/HrishabhCodes)