Every great story starts with a choice
Nine years of building software. Started as a teacher. Still think that way.
Full stack by practice, not just title. The frontend, the backend, the infrastructure — all of it, because understanding the whole system makes you better at any part of it.
Currently learning AI engineering properly — not just using the tools but understanding what they actually do. This portfolio and the chatbot powering it are part of that.
More being built. Always.
Started as a teacher. Still thinks that way — about clarity, about making hard things understandable, about leaving things better than they were found. Nine years of engineering built on top of that.
Currently moving into AI. Curious by nature. Homelab in progress. Anime nerd. Always building something.
These are not metrics. These are victories.
Large-scale healthcare platform — Micro Frontend architecture, Vue, ASP.NET microservices, GraphQL orchestration layer.
The technical side was manageable. The harder part was keeping standards consistent across 15 different teams all building on the same foundation, with different timelines and priorities.
Reduced API call volume by 50% — from around 1 million daily calls down to 400K — through GraphQL and API layer optimization. Not a rewrite. Incremental improvements to an existing system.
Also led the implementation of an AI-powered search feature — from proof of concept through to production. Designed the rollout, benchmarked at each stage, and shipped it safely.
Sole developer on an internal tooling project for a large technology company. Owned everything — architecture, development, deployment, and stakeholder communication.
Built the first Angular and Python application for the team. Integrated with internal dashboard tooling, automated reporting workflows, and optimized database queries that were slowing down data loads for analysts.
Also led a full security compliance upgrade — from Level 3 to Level 4 — as the only engineer on the project. That one took longer than expected, but it got done.
Billing platform enabling members to make premium payments securely. Payment-critical systems require a different standard of care — reliability, security, and no ambiguity in what the system should do.
Internal platform allowing engineering teams to provision and host their own databases and servers. Developer tooling done well makes everyone else faster. That was the goal here.
Worked across a large-scale Medicare and Medicaid platform with 15 engineering teams. Reduced API traffic from 1 million to 400K daily calls through GraphQL optimization and API layer improvements. Led an AI-powered search feature from proof of concept to production deployment.
Designed and led the full rollout of an AI-driven search capability. Built the phased deployment strategy, benchmarked performance at every stage, and ensured safe adoption at scale. First production AI feature in the domain.
Sole engineer on an internal tool for a large technology company. Owned architecture, development, and deployment. Built the first Angular and Python application for the team, automated reporting workflows, and led a security compliance upgrade from Level 3 to Level 4.
Built this portfolio and the RAG-powered chatbot from scratch. Self-hosted on a Raspberry Pi 5 using Docker Compose. The chatbot answers questions grounded strictly on resume and context data — no hallucinations, no invented information.
Not a checklist. A record of things used under pressure.
Things used seriously over the past nine years. Some daily. Some when the project called for it. All of it shipped to production at some point.
Angular has been the longest relationship — RxJS, NgRx, component architecture, and performance optimization at scale. React and Vue followed. Each has its place.
TypeScript, Webpack, Storybook, WCAG accessibility compliance — the details that separate a working application from a maintainable one.
ASP.NET Core for high-volume microservices. Python for automation, data tooling, and increasingly AI work. Node.js where it fits. GraphQL as an orchestration layer — and the experience of optimizing it when it becomes the bottleneck.
Azure is home territory. GCP from a major consulting engagement. Docker, Kubernetes, CI/CD pipelines — not just set up once but maintained and improved over time.
LangChain, RAG pipelines, OpenAI API, Pinecone. Currently the most active area of learning — building things, not just reading about them.
MIT Professional Education — AI: Implications for Business Strategy University of Helsinki — Elements of AI Adobe Journey Optimizer
Step forward. Make contact.
If something here is interesting or you have a question — reach out. Always happy to have a real conversation.
AI-MC can answer questions about experience and background. For everything else the links below go to the right places.
Have questions? Ask AI-MC — the button is always in the bottom right.
The side of Mahesh the resume never shows
Since you made it here —
Early mornings, consistent effort, kickboxing when the week has been rough. Staying active keeps everything else in balance.
New trails, new cities, new places. The best thinking happens away from a screen.
Deeply, unashamedly, a huge fan. This portfolio is built around it deliberately — each chapter a different universe. The storytelling, the art direction, the way a good arc makes you feel something.
Current favorites: Hunter x Hunter, Hajime no Ippo, One Piece. Always open to recommendations.
Something being learned slowly — not as a career change, just genuine interest in the craft. How scenes are directed, how motion carries emotion. It has been on the list for a while.
The homelab. This portfolio. The chatbot. Side projects that exist because building them was interesting, not because anyone requested them.
This is where the real learning happens.
Hey there! I'm AI-MC, Mahesh's digital assistant. I'm trained on his resume and experience — nothing more, nothing less. Here are some things you can ask me: