From college contests to Big Tech in 6 months

Karim El-Zawawi walked into Amazon Jordan three years out of university and stayed for three. He’s now at Careem after a stop at Atypon. He insists the walls between an Arab fresh graduate and a Big Tech offer are mostly imaginary — if you know what to focus on, three to six months is enough.

When students ask Karim El-Zawawi the one thing they should have done differently in college, he answers without hesitation: everything that happened outside the lecture hall. The degree got him to the gate. ICPC programming contests, side projects, and self-study were what carried him through it.

His path is honest in a way most LinkedIn bios aren’t: third place in the Jordan ICPC during university, three years at Amazon Jordan, two at Atypon (a Wiley brand), then a move to Careem on the captain-routing system — the algorithm that decides which driver gets your order, and which streets they take. Along the way, he built a paid course on Elastic Search for educative.io. He doesn’t pretend his career is a magic trick. He just refuses to dress it up as one.

The biggest myth: that Big Tech is impossibly hard

Karim is direct: most people who think they can’t get into Amazon, Google, or Careem are building walls that aren’t there. Three to six months of focused prep is enough for a junior or fresh grad. He’s watched it happen to candidates around him, repeatedly. The day-to-day tasks at Big Tech aren’t harder than at a small company — in many ways, they’re cleaner.

What stops people is the mental image: a hoodie-wearing genius who knows seven programming languages and just rolled out of MIT. Real candidates don’t look like that. Real interviewers don’t care about that.

What Big Tech actually screens for

After years of giving interviews, Karim is clear that Big Tech filters on three things — not the specific tech stack on your CV:

1. Problem solving. Algorithms and data structures. Open LeetCode (he prefers LeetCode over Codeforces for interview prep), and start solving by yourself. Not watching solutions — solving. The muscle is built by trying, getting stuck, and pushing through.

2. System design. Once you cross the junior level, every senior interview revolves around this. “Design Uber from scratch.” “Design Amazon checkout.” They’re not testing your knowledge of a framework. They’re testing whether you think about scale, traffic, edge cases, and how the system breaks.

3. Behavioral skills. Ownership and communication. Can you take a vague problem and run with it? Can you walk an interviewer through your thinking out loud, not just deliver the answer?

Junior vs. senior: the real difference

A junior gets a task, hands it in 80–90% done, and a senior reviews it. A senior, on the same task, automatically asks bigger questions: What if user count doubles? Is this code going to be bad in a year? Are we adding tech debt? Does this approach break when the team grows?

It’s a posture, not a title. You can practice it right now. Read your own code as if a future team will inherit it — because they will.

How to actually learn programming (his version)

Karim is uncompromising about how juniors should learn. He calls out the 40-hour video courses by name: they don’t work. You don’t learn programming by watching it.

His method:

Take a one-to-three hour crash course on a topic. Then close it. Open a blank editor. Build something with that topic — from scratch, no tutorial open. Reading is 10% of learning. Doing is the other 90%.

Focus on concepts, not syntax. Every language has different syntax, but the concepts repeat: data structures, scoping, concurrency, transactions. Learn the why, and the “what” becomes a Google search.

Look inside the tools you use. Don’t just learn how to use a database — learn how a database is implemented internally. Storage engines, indexes, transaction logs. This is what separates an engineer who solves real production problems from one who copies snippets from Stack Overflow.

Don’t specialize too early

A common mistake: a fresh graduate brands themselves as “backend developer” or “frontend developer” from day one. Karim moved between Java at Amazon, Python at Atypon, and Go at Careem. Big Tech doesn’t hire “Java backend engineers” — they hire engineers who can pick up Java in a week and ship.

The skill they actually buy is the ability to learn quickly. If you trained for that, switching from Spring to Django is one weekend of study. If you trained only for Spring, switching is six months and an identity crisis.

AI as an engineer: helpful, not an identity

Karim uses Cursor and ChatGPT daily. He’s not anti-AI — he’s anti-laziness. AI as an engineering tool: powerful, save hours, integrate into your workflow. AI as a career identity (“I want to be an AI engineer”): mostly a marketing label that’ll age badly.

The skill worth picking up is integration: how to wire AI into the codebase you already maintain. Concepts like RAG (Retrieval Augmented Generation), MCP (Model Context Protocol), and agentic workflows are where real engineering value lives. Switching careers to be “an AI person” isn’t the move — learning to make AI useful inside your existing engineering work is.

A 90-day plan, if you want one

If you’re a fresh grad or junior reading this and wondering where to start: pick LeetCode, commit to it daily for three months, build one side project that reaches real users (even if there are only ten of them), and start running through System Design Interview videos on YouTube once you’ve cleared the basics. That’s the entire blueprint. There’s no secret you’re missing.