Revisiting My Master's Degree: Six Years Later

In the fall of 2018, I started taking graduate courses at UC San Diego while working full time as an electrical engineer in the defense industry. By spring 2020, I had completed a master's degree in electrical engineering: twelve courses spanning VLSI design, communication circuits, machine learning, antennas, control systems, and motor drives. At the time, I wasn't entirely sure what I'd do with it. Now, six years out, I'm starting a series to revisit each course and evaluate what turned out to matter in my career since and where the technology topics have progressed since.

This is an attempt to help me reflect and understand. I want to revisit what I expected to learn versus what I actually learned, which concepts have stayed relevant in my work, and which felt important then but disappeared from view. I'm also curious about how the field itself has shifted. For example machine learning in 2019 looked very different than it does now. RF circuit design principles are relatively stable, but the applications and constraints have evolved.

Why I Did It

I'd always wanted to pursue a graduate degree. I'm curious by nature, and I wanted to keep learning beyond what my undergraduate education and early work experience had given me. I knew there were entire domains of electrical engineering I'd barely touched, and a graduate program felt like the right way to get exposure to those topics in a structured, rigorous way. I also believed that deeper technical knowledge would help me advance my career, even if I wasn't sure exactly how. I wasn't aiming for a particular role, but I knew I wanted to expose and challenge myself in specializations across different areas.

The practical reason was simpler: my employer offered a tuition reimbursement benefit, and the in state tuition at UCSD was below the maximum benefit. The financial risk was zero, and the opportunity felt rare given where I was in my career. It was early enough that I still had the bandwidth to take on something this consuming, but no doubt I was naïve to just exactly what I was getting into. My main thoughts were the degree might position me either for more specialized design work or technical leadership down the line.

There was also a less tangible motivation. I'd been working in digital design for a few years, mostly focused on FPGAs and embedded systems, but I'd always been interested in RF. RF design seemed more challenging, harder to understand, harder to build intuition for, and harder to get right. The physics mattered more. Parasitics and electromagnetic effects that you could mostly ignore in digital designs became first-order concerns (though nowadays high speed digital has many of the same concerns). I wanted to understand that world, even if it wasn't what I was doing day-to-day. UCSD's RF program also had strong ties to Qualcomm, and the faculty were working on problems that mattered in real wireless systems. That practical grounding appealed to me. I didn't just want to study abstract theory, I wanted to learn from people who were connected to the challenges that industry actually cared about.

What It Was Actually Like

The first quarter was a rude awakening. I took three courses: VLSI design, solid state electronics, and digital image processing. All three were harder than anything I'd done in undergrad, and I hadn't anticipated how much time I'd need just to keep up. I was still adjusting to full-time work and a full-time course load, and I hadn't yet developed the discipline to study efficiently in the narrow windows between work and sleep. I spent weekends trying to catch up on problem sets, but often times getting stuck at points where my notes were not sufficient. I was also still taking handwritten notes on paper, while nearly all my classmates were using iPads. I didn't realize what a disadvantage that was until a few weeks in. The lectures were all given with PowerPoint slides, and everyone else could just annotate directly on the digital files. I was frantically copying diagrams and equations by hand, losing time and clarity in the process. My notes from that quarter are a mess.

By the second quarter, I'd figured out more of a rhythm. I started blocking off specific hours for coursework and treating them as non-negotiable. I also started taking the day before exams off from work, which made a huge difference. It gave me time to review without the mental fatigue of a full workday. More than anything, I just got better at being an effective student again. I'd been out of coursework for three years, and I'd forgotten how to study efficiently.

The content itself was unrelenting. Graduate courses don't scaffold the way undergraduate courses do. They assume you can fill in gaps on your own, and they move quickly through foundational material to get to the interesting edge cases and tradeoffs. Communication circuit design required me to think about noise, linearity, and frequency-domain behavior all at once. Random processes forced me to become fluent in probability theory I'd barely been exposed to in undergrad. Every course demanded something different, and that variety kept me engaged even when I was exhausted.

By the end of the program, I felt sharper. Not just in technical knowledge, though I'd certainly expanded my toolkit, but in how I approached problems. I was better at decomposing complexity and determining sound approaches across domains. I also felt more confident speaking about topics outside my immediate area. I could hold my own in conversations about antenna design or machine learning architectures (at least back then) even though I'd never worked directly on those problems. That breadth turned out to be more valuable than I'd expected.

What I Hoped to Gain vs. What Actually Mattered

I went into the program with a few concrete goals. I wanted to expand my EE knowledge beyond the digital design work I'd been doing professionally. I'd taken one RF course in undergrad, but that was years ago and I'd spent all my professional time in digital design. I wanted to be around people who were accomplished academically and held themselves to high standards. And I thought that learning RF design would make me better at digital design, that understanding analog systems would give me better intuition for what was happening at the physical layer.

Those things did matter, though not always in the ways I expected. Being in classes with people who took the work seriously raised my own standards. The rigor was contagious. I found myself caring more about understanding concepts deeply rather than just getting through problem sets. The RF coursework did improve how I think about digital designs, particularly around signal integrity, noise coupling, and timing. When you've worked through the math of how signals behave in transmission lines or how noise propagates through amplifier stages, you develop better instincts for what matters in high-speed digital circuits.

But the thing that's stayed with me most is something harder to quantify: I got better at learning complex material and figuring out how to apply it to real problems. Graduate courses don't hold your hand. They give you frameworks and expect you to fill in the gaps, to figure out which parts matter for your specific situation, to know when the approximations break down. That process of taking abstract concepts and grounding them in concrete constraints, that's what I do constantly now. Whether I'm evaluating a new design approach, debugging a system-level issue, or just trying to understand a technology I haven't worked with before, I'm using that same muscle.

The breadth helped too. I can move between conversations about antenna design, machine learning, control systems, power electronics, and contribute something useful even though none of those are my primary focus. That's valuable in ways I didn't anticipate when I started. It's not about being an expert in everything, it's about having enough context to bridge different specializations and see connections that aren't obvious from inside one domain.

The Curriculum: What I Actually Took

The program required twelve courses over six quarters. Here's what I ended up taking:

Fall 2018: VLSI Digital System Algorithms and Architectures, Solid State Electronics I, Fundamentals of Digital Image Processing

Winter 2019: Communication Circuit Design I, Security of Hardware Embedded Systems

Spring 2019: Communication Circuit Design II, Machine Learning for Physical Applications

Fall 2019: Power Amplifiers for Wireless Communications, Antennas and Their System Applications

Winter 2020: Random Processes, Mathematical Topics for the Master's Comprehensive Exam

Spring 2020: Linear Control System Theory, Motor Drives

Looking at this list now, a few things stand out. First, the program was heavily tilted toward analog and RF design: four courses in communication circuits, power amplifiers, and antennas. This was deliberate on my part. Second, the machine learning course happened in spring 2019, right before the field exploded into mainstream consciousness. The course focused on physical applications and classical methods: Gaussian processes, support vector machines, kernel methods. It barely mentioned deep learning. Third, I took a random processes course and a dedicated math review course back-to-back in my fifth quarter. Those two courses, more than any others, gave me the mathematical fluency I needed to work through the later material.

The courses weren't perfectly applicable to my day-to-day work at the time. I was doing digital design for embedded systems, mostly dealing with timing constraints and FPGA architectures. The VLSI course was directly relevant, and the hardware security course gave me useful context, but most of the other courses felt orthogonal. That worried me initially. I questioned whether I was just accumulating knowledge that would sit unused. But over time, the connections emerged. The intuition I developed in one domain turned out to be transferable to others.

How This Series Will Work

I'm going back through each course, one at a time, and asking a few questions:

1. What were the core ideas? Specifically the 3-5 concepts that defined the course and have stayed with me.

2. What has changed since 2018-2020? Some fields move slowly; others transform completely. I want to understand how each domain has evolved and whether the foundational concepts still hold.

3. What actually mattered? Which ideas from the course have I used in practice, directly or indirectly? Which felt important at the time but turned out to be peripheral?

4. What would I prioritize if I were taking it today? Knowing what I know now, which topics would I dig into more deeply, and which would I skim?

The point of this series is not to rehash the entire curriculum or pretend to be an expert in every topic. These posts are meant to be reflections, not tutorials. I'll share what I remember, what I've used, and what I've since learned. I'll be honest about the gaps: places where I never fully understood the material, or where I've forgotten more than I've retained.

Each post will also include a brief look at the field itself: who's working on these problems now, what the open questions are, and how the landscape has shifted. I'm curious whether the problems we studied six years ago are still relevant, or whether the field has moved on.

Why Revisit This Now?

It's been six years since I finished the program. That's long enough for the immediate details to fade but not so long that I've lost the thread entirely. I want to look back while I still remember what it felt like to struggle through a problem set or finally understand a concept that had been opaque for weeks.

The timing also matters because the field has changed in visible ways. Machine learning has gone from niche academic topic to omnipresent tool. RF and wireless systems are now central to 5G infrastructure and IoT deployments in ways they weren't in 2019. Hardware security went from specialized concern to board-level priority.

There's also a selfish reason: I want to see what stuck. I invested two years and significant energy into this degree, and I want to know which parts of it are still with me. Memory is unreliable, and it's easy to overestimate how much you've retained. Going back through the material, even if only at a high level, forces me to confront what I actually know versus what I think I know.

Finally, I think there's value in documenting this kind of reflection. Most discussions about graduate education focus on whether to pursue a degree in the first place, not on what happens after. People want to know if it's "worth it." In reality the utility isn't linear, and it's not always obvious until years later. I want to capture that complexity.

Who This Series Is For

First and foremost this series is for me to reflect. Alternatively, it's for people who are thinking about graduate education and want a ground-level view of what it's actually like. It could also be for people who completed a technical degree years ago and are curious what a masters program looked like between 2018-2020. If you took courses in control systems, RF design, or machine learning in the late 2010s, you might find it useful to see how those ideas have aged and what the field looks like now.

Lastly, it could be for working engineers who are trying to decide where to invest their learning time. You don't need a master's degree to build expertise, but understanding what a formal program actually covers, and what it doesn't, can help you calibrate your own learning path. If you're self-studying random processes or communication circuit design, these posts might give you context for why certain topics get emphasized and others get skipped.

What's Next

The first post will cover VLSI Digital System Algorithms and Architectures, the course I took in my very first quarter. From there, I'll work through the rest of the curriculum in roughly chronological order, with some exceptions. I'll group the communication circuit design courses together since they form a natural sequence, and I might jump around based on what feels most relevant to write about at the time.

Again the goal isn't to be exhaustive. The goal is to reflect honestly on what the degree gave me, what it didn't, and how the ideas have evolved. If you're here for that, I hope you'll find something useful.