LearnLog
Who Am I?
Hi! I'm Rohit Tyagi, a software engineer with 6+ years of experience building products that scale. My journey started in frontend engineering, crafting robust web and mobile applications, architecting scalable systems, and leading teams at Mosaic Wellness as an SDE3.
But here's where it gets interesting: I've pivoted into the world of AI.
Today, I'm building at Zonko, a frontier AI lab working on cutting-edge problems. My focus has shifted from pixels and components to deep learning and agentic AI systems. I'm fascinated by how we can build intelligent systems that don't just respond, but reason, plan, and act autonomously.
The transition from frontend to AI wasn't a leap, it was a natural evolution. The same principles that made me effective at building user-facing systems, attention to detail, system thinking, and user-centric design, now apply to building AI systems that millions will interact with daily.
What Is This?
This repository is my learning journal, a curated log of everything I'm discovering in AI, deep learning, and agentic systems. It's also a reflection of my frontend roots, documenting the patterns, architectures, and insights that shaped my engineering journey.
Software engineering is dynamic. AI is moving even faster. This is where I capture that evolution: experiments, breakthroughs, failures, and the compounding knowledge that can't be copied.
Where I've Built
Current: Zonko
Building AI-native products that could reach billions. We're shipping fast, watching what spreads, and compounding taste until something breaks through.
Previous: Mosaic Wellness
Built and scaled multiple health & wellness platforms:
- Man Matters Android | iOS
- Our Little Joys Android | iOS
- BeBodywise Android | iOS
- MediConnect Healthcare platform
Earlier: Navis & Anar
- Navis: Built logistics and terminal management apps (nReefer, Pinning, Terminal Visits)
- Anar: B2B marketplace platform and mobile apps
Resources I Learn From
- React Internals: Which useEffect Runs First? Deep dives into framework internals
Let's Build Something
Interested in AI systems, deep learning, or building products that scale? Let's connect: