My Story 📖
From learning C in a college classroom to founding TurboChat — Rakesh Jha on building AI products, experimenting with LLMs, and shipping things that matter.
I'm Rakesh. I build stuff with code and run TurboChat. Here's how I got here.
School and College
Grew up in New Delhi. Normal middle-class life, nothing unusual. School was fine, I wasn't特别 good or特别 bad at anything. Didn't touch code until I joined Noida International University for CS.
First year was mostly C and Java in class. I did the assignments, showed up for exams. It wasn't until later that I started coding for fun — building random things, breaking them, figuring out why they broke.
First Job
April 2024, I joined Alfastack Solutions. My title was Full Stack AI Engineer but honestly I was just figuring things out as I went. Built an AI pipeline in three weeks that connected Frappe with LLMs. It cut manual work by 40%. That's when I realized code can actually move numbers.
Then at AiSolv, I built automation agents — lead scrapers, workflow stuff, whatever saved people time.
Chakra Labs
At Chakra Labs, I'm on the Dojo project. It's a platform for RL researchers. I build the frontend — dashboards, training viz, that sort of thing. The requirements change a lot because research moves fast. You just have to keep up.
Starting TurboChat
I started TurboChat in early 2026. The idea was dead simple — paste a URL, get a chatbot. No setup, no infrastructure. It just works.
It's at 600+ signups now. What I learned building it is that people don't care about your tech stack. They care if the thing solves their problem.
Random Stuff I've Built
A blockchain document verifier. A voice AI agent that runs in the browser. An image-to-dataset tool for ML. A developer memory layer called Shadow. A Go HTTP server from scratch. A GEO tool that tracks AI citations.
I just build whatever seems interesting at the time. Some of it works, some doesn't. I learn either way.