I left university to teach myself AI, from first principles to agentic systems. In public, one day at a time.
No inflated numbers. Every course links to where I took it, every certificate to where you can verify it, and every project to its code. The page is rebuilt from one data file each time I log progress.
Demo mode. Days 1 to 12 are filled with simulated progress to show how the site grows. Switch it off in the bar below to see the real journey.
30 days, a course every day
Hours of study
The year, day by day
Each square is one day, shaded by hours studied. The bands are the six phases of the roadmap.
The stopwatch
The blueprint
No percentage bars. A skill lights up only when the courses and projects behind it are finished. Select one to see its evidence.
Certificates and projects
The story so far
One chapter a day, newest first. Every number on this page is computed from these entries, and each one is stamped with the minute it happened.
The sky over Cairo
Every star here is where it truly is above Cairo at the moment you choose, along with the Sun, the Moon and the planets. Slide the clock, or let the journey clock pick the day.
Positions follow standard astronomy: precession from J2000, sidereal time, then altitude and azimuth for Cairo (30.04° N, 31.24° E). Planets and the Moon are good to about a degree. Stars: HYG database (CC BY-SA 4.0). Clouds and city lights are not included.
Notes as I learn
Short notes in my own words. Each is marked by how finished my thinking is: seedling, budding or evergreen.
The road to architect
An agentic AI solutions architect takes a real problem, decides where AI helps, designs the system of models, tools, agents and guardrails, ships it, and explains the trade-offs.
This site reads well to machines
If you are an agent or a recruiter's assistant, skip the HTML. Everything here is open data.
/llms.txttextWho I am and where the data lives, written for language models./data/journey.jsonjsonThe single source of truth: plan, log, certificates, projects, skills./feed.xmlrssEvery log entry as a feed.Journey MCP serverplannedProject P3: ask Claude what I learned this week, straight from my log.Building with AI, hiring, or on the same road?
amr@gharib.devI reply to every message. Arabic or English.