Why I Built This Roadmap — And Why I'm Updating It Every Single Day
I have been in XR and software development for 13+ years. Over the last few years I have watched AI go from something I had to wire together manually with five different APIs — Dialogflow, Watson, AutoML, Parallel Dots — into something that now runs entire workflows on its own. That shift changed what companies expect from developers, and it changed what it means to be a senior AI engineer in 2026.
I started getting questions. From developers who wanted to move into AI roles but did not know where to start. From people who had done a few courses but still could not get past the resume screen. From engineers who knew Python but had no idea what an AI Architect actually does day to day.
So I built this roadmap. Not a generic one. Not a list of YouTube videos. A structured 80-day plan that takes you from writing clean Python to designing production-grade Agentic AI systems — specifically tuned for BFSI — banks, asset managers, wealth management firms, insurance companies, and fintech. The projects, the datasets, the compliance angles — all of it is built around what financial services companies are actually hiring for in 2026.
What makes this different
Most roadmaps tell you what to learn. This one tells you what to build. Every single phase has a real project attached to it — not a toy example, not a tutorial follow-along. Something you can put on GitHub, write about on LinkedIn, and talk about in an interview.
The 8 phases are sequenced deliberately. You cannot jump to building Agentic AI systems without understanding how embeddings and vector search work. You cannot design AI architecture without knowing how models behave in production. The order matters — and it follows the same progression that senior AI roles actually test for in interviews.
How this page works — updated every day
This is not a static page. I am running through this roadmap myself and publishing a dedicated blog post every single day starting from Day 1. Each post goes deep on that day's topic — what I actually did, what broke, what I learned, and how it connects to real hiring requirements.
The interactive roadmap above shows all 8 phases. Click any phase to see the full curriculum, tools, build projects, and recruiter keywords. As I publish each day's blog post it will link directly from here — so you can follow along day by day, or jump ahead to any phase you are interested in.
Who this is for
If you are a developer with at least 1–2 years of coding experience who wants to move into an AI Architect or senior AI Engineer role — this is for you. You do not need a machine learning background to start. Phase 1 assumes you can write Python, that is it.
If you are already doing some ML work and want to level up to Agentic AI, system design, and MLOps — jump to Phase 4 or 5. The phase cards below have enough detail to figure out where you belong.
And if you are a hiring manager or recruiter reading this — every project in Phase 6 of the portfolio is a real, deployable system. Not a notebook. Not a Colab demo. An actual production-ready build.