🎓 My Georgia Tech OMSCS Journey: 10 A’s, Countless Lessons (and a Lot of Late Nights)

🔥 CS 6515: Graduate Algorithms

• Difficulty: 5/5
• Effort: 40 hrs/week
• Satisfaction: 5/5

This course was the final boss of OMSCS — I dedicated an entire semester to it.

The professor said, “I'll teach you Dynamic Programming,” and he did. But the exam had a wild DP twist I hadn’t seen before. I stayed calm, applied the patterns I'd practiced, and cracked it. That one question gave me more joy than most things that semester. During the final exam, I got COVID, and it was a nightmare. I was bedridden, but I still managed to finish the exam. I remember constantly coughing while giving the exam. It was brutal.

I practiced relentlessly, joined weekly group calls, and solved endless problems from the book. This course humbles you, sharpens you, and rewards deep focus.

🗣 Toughest in the program, but extremely rewarding. A must-do.

🧠 CS 7641: Machine Learning

• Difficulty: 5/5
• Effort: 40-45 hrs/week
• Satisfaction: 4.8/5

No spoon-feeding here.

This course wants you to understand, not just use libraries. We dove into the math, assumptions, and internal workings of every algorithm. Projects were open-ended. Reports were intense. It felt brutal at times — but when you get it, you get it for life.

I had to work double to understand the math behind algorithms. I had to read papers, watch videos, and even ask for help from my peers. It was a grind, but it paid off.

I worked on two real datasets, wrote detailed reports, and debugged for hours. And yes, you will cry over grading rubrics.

🗣 One of the heaviest workloads. Also one of the most respected.

🧩 CS 7637: Knowledge-Based AI

• Difficulty: 3/5
• Effort: 12 hrs/week
• Satisfaction: 4.5/5

My very first OMSCS course (paired with Databases). People say it's easy — I say it’s conceptually light, but the RPM project still takes effort.

The lectures were clear, and Prof. Joyner was excellent. I even built a LaTeX template for submissions: jdf-latex, which got recognized by other students (and even in homework feedback!).

🗣 Great intro course, especially if you want to ease into the program.

🌐 CS 6250: Computer Networks

• Difficulty: 3/5
• Effort: 10–15 hrs/week
• Satisfaction: 4/5

Solid and practical. Cleared up so many core networking concepts: subnetting, routing, and autonomous systems. The simulations made things click, and this course ended up being quite helpful in tech interviews too.

🗣 Well-structured, practical, and useful.

🧱 CS 6400: Database Systems Concepts and Design

• Difficulty: 3/5
• Effort: 12 hrs/week
• Satisfaction: 4.5/5

Paired this with KBAI. Excellent intro to databases. We designed schemas, wrote tons of SQL, and built working systems. But what stood out most was the professor’s office hours — loaded with off-syllabus gems.

🗣 Ideal for brushing up real-world DB skills.

⚖️ CS 6603: AI, Ethics, and Society

• Difficulty: 3/5
• Effort: 8–10 hrs/week
• Satisfaction: 4/5

Looking at AI through an ethical lens was refreshing. We explored bias, fairness, surveillance, data justice — and how all of these actually affect real ML systems. Great mix of philosophy and application.

🗣 An easier course, but deeply thought-provoking.

🤖 CS 7638: Artificial Intelligence for Robotics

• Difficulty: 4.2/5
• Effort: 15–18 hrs/week
• Satisfaction: 4.2/5

People say it's “easy” — not entirely true. The concepts are clear, but getting the code right takes patience. Lots of test cases to pass. That said, it’s a great way to understand localization, planning, and control.

🗣 Good balance of challenge and fun, especially if you like hands-on work.

💹 CS 7646: Machine Learning for Trading

• Difficulty: 3.9/5
• Effort: 12 hrs/week
• Satisfaction: 4.5/5

Joyner again! This course blends finance and ML beautifully. You’ll build trading strategies, learn how to analyze past performance, and apply regression, reinforcement learning, and more.

This was one of the most fun and practical courses — especially if you’ve ever thought of dabbling in quant stuff.

🗣 Superb course. Lots of insights into algorithmic trading.

📊 CSE 6242: Data Visualization and Analytics

• Difficulty: 3.8/5
• Effort: 10–12 hrs/week
• Satisfaction: 4.3/5

Here’s what made this course awesome: every assignment used a different tech stack. One had me using Spark, another Hadoop, one on Azure. There was no ChatGPT back then — I figured it all out manually. 😅 This subject trained you to pick up unknown technologies in every homework, one time I was writing hadoop cluster in Java, and the other week, reading a data through spark.

Beyond tech, the real power was in learning how to tell stories with data. Choosing the right visualization, the right framing — it stuck with me.

🗣 A practical, tool-heavy course that pays off in the real world.

🧑‍💻 CS 6750: Human-Computer Interaction

• Difficulty: 2.8/5
• Effort: 8–10 hrs/week
• Satisfaction: 4.5/5

This course changed how I think about software design. It’s not just about building features — it’s about making them usable, accessible, and delightful. We did user research, prototyping, and real UX work.

I picked up Airbnb and Github as case studies, and designed 10-20 mockups for each. I remember sharing the Github case study with someone who used to work at Github, and later it was part of Github's review system.

Oh, and you also have to get a certified human behaviour researcher certificate in order to conduct user studies. It’s a bit of a hassle, but it’s worth it.

🗣 Creative, thoughtful, and very different from the rest of the program.

🧘 Final Thoughts

Specialization: Machine Learning (primary), with additional coursework in Interactive Intelligence (AI). Though the degree formally mentions Machine Learning, I completed courses spanning both tracks.

Was OMSCS hard? Hell yes.

Worth it? 1000%.

Courses like Graduate Algorithms and Machine Learning took everything out of me. Others like KBAI, AI Ethics, and DataViz helped me breathe and grow in unexpected directions.

And let me say this clearly — I didn’t pick courses just to get by. I picked what I actually wanted to learn. I didn’t look for "easy A’s" like many people do just to cruise through the program. I chose topics that challenged me — even if it meant pulling crazy hours, facing burnout, and questioning life choices at 3 AM.

I was spending 12–14 hours a day working across jobs and courses. There was no time to pause. I had to let go of almost everything — family time, friendships, outings, festivals — to keep this going.

People saw a 4.0 GPA, but they didn’t see the sacrifice behind it. It drained me to the core — but in the end, I crushed it: 10 courses, 10 A’s, and a perfect 4.0 GPA.

I didn’t attend a single wedding or family function for 2.5 years. A friend once asked me on a call, “Where have you been?” I told him, “Himalayas. Went to take sanyas.” (Translation: I went to the Himalayas to renounce everything — like a monk.)

We both laughed. But honestly? It wasn’t far from the truth.

While all this was happening, I was also handling 2–3 client-facing projects at my job — juggling between Python, JavaScript, Node.js, Django REST Framework, and even DevOps work with Terraform and AWS.

At Georgia Tech, it was Hadoop one day, Spark the next. The amount of context-switching I did became second nature. These days, even when I get pulled into unfamiliar tools or vague bug reports, I’m usually able to piece things together quickly — not because I know everything, but because I’ve trained myself to adapt fast and think clearly under pressure.

People are often surprised at how calmly I can trace issues — whether it's a teammate or a client — but that's just the result of years of repetition, context-switching, and problem-solving under fire.

And truthfully — I wasn’t the only one struggling.

I saw people burn out, break down. I saw relationships strained, even divorces. This program is no joke if you’re balancing a full-time job. It demands everything. You sacrifice time, mental energy, and sometimes emotional bandwidth. You need support. You need a reason. And you need to really want it.

Many people take 5+ years to finish OMSCS, taking breaks along the way — and that’s completely valid. But I’ve always been wired a bit differently.

My thinking was: if it’s going to hurt, let it hurt now — not stretch that pain for 5 years. I wanted to get through the storm, not set up a tent in it. So I fast-tracked it. Full throttle. No looking back.

There were days I’d sit in front of the screen and just feel empty. Assignments pending, Slack messages piling up, eyes burning from lack of sleep — and yet, I had to keep going.

Sometimes it felt like I was sprinting on a treadmill that wouldn’t stop. No applause, no weekends, no certainty — just sheer willpower. But maybe that’s what growth looks like in real life — not in quotes, not in movies — but in showing up when nobody’s watching.

One of the blog posts I once wrote quoted The Martian, and it stuck with me throughout this journey:

“At some point, everything’s gonna go south and you’re going to say, this is it. This is how I end. Now, you can either accept that, or you can get to work. That’s all it is. You just begin. You do the math. You solve one problem and you solve the next one and then the next. And if you solve enough problems, you get to come home.”

I had this moment during one of the toughest and most horrifying subjects — Graduate Algorithms. I caught COVID, along with my family, and suddenly, it became an emotional battle too. But I remembered this quote.

That's exactly what I did.

I didn't have answers. Just grit.

I solved one bug. One assignment. One test. One project. Then the next. And eventually — I got to come home.

There’s a line I wrote once that still defines how I operated during this time:

“I ran like a cheetah that knows it doesn’t get to eat if it doesn’t catch the prey. No second chances. No tomorrow. Just now.”

That was my mindset for 2.5 years.

And throughout all of it — whether I was pushing code, climbing a mountain, or debugging something at 2 AM — there was one line I kept telling myself:

“Bas ho gaya.” (Just a little more. Almost there.)

It became my mantra. Not because it was easy. But because I refused to stop until it was.

Would I do it all over again?

Probably not.

But would I trade the lessons, the growth, the confidence for anything else?

Never.

I wasn’t lucky — I was tired. But I didn’t stop. My only escape was my ambition.

The degree gave me confidence, clarity, and resilience. Now it’s time to use it.

You can check out the projects I did throughout the program at sanyamkhurana.com. If you're thinking of diving into OMSCS or just want to swap stories, feel free to reach out.

I’m not a genius. I’m just a guy who wanted to learn. I didn’t have a plan. I didn’t have a roadmap. Just a promise to myself that I wouldn’t quit.

“No shortcuts. No hype. Just consistent work — and a quiet promise to finish what I started.”

CuriousLearner; now a Yellow Jacket became a ramblin' wreck from Georgia Tech and a hell of an engineer.