PhD hot takes as a 2nd year PhD student

I filmed a TikTok video discussing my PhD hot takes and thought it would make for a good blog post (I’ve not updated this blog in AGES). So, let’s get into it!

1 – It’s ok to work on the weekends

I heard a lot about the poor work life balance within PhD students, working 60 hour weeks and working weekends. I’m still big on work-life balance but I think in some ways, we have gone the other extreme of strictly working 9-5 hours for a PhD.

As much as the working hours fit a 9-5, the kind of work you do on a PhD is very different to corporate. There’s a lot of time spent thinking, reflecting, reading and writing ideas. Inspiration often strikes at the weirdest times (I’m a composer and I get composition ideas at 11pm). So, when I had a nice idea for a research project on Saturday at 4pm, I’m going to write it down. I don’t do a full day of work on the weekends, but I definitely do some planning and organising when I feel like it.

I’m also the kinda person who will forget to do something if I don’t do it there and then. If on Sunday, I remember that I need to do a supervisor meeting form or book a library room, I would rather get that done then, over waiting until Monday just to say I have “work-life balance”. Schedule sending email is a power tool for this.

Work-life balance is less about working the least hours possible but more about how you allocate those hours and whether you feel balanced in the way you work. Working on the weekends doesn’t automatically mean poor work-life balance, especially if maybe there’s some weekdays where you don’t do as much work. As long as the work gets done and you’re looking after yourself, I have nothing against working on the weekend (and I’ll probably keep doing it from time to time).

2 – You don’t need a whole productivity system

Notice how productivity advice for PhD students is rare? I think it’s because the PhD just can’t be systemised in the same way as an undergrad degree or a masters. There are some principles that help, such as time-blocking but the way you work as a PhD student doesn’t really fit a traditional productivity system. You can’t measure your tasks by how long they take because some of the most important tasks take hours (reading papers) but something like replying to emails can be done in a few minutes. The time to value ratio is completely skewed on a PhD.

I trialled a bit with different productivity systems, but I mostly found that it was a distraction from getting the actual work done. The one practice that I have stuck to is daily logging. Writing down what I did each day on the PhD has helped so much with seeing the progress I’m making over a period of time. It’s not a hard one to do, and it doesn’t take long but it makes a huge difference. I would argue it’s more important than setting up a fancy calendar or having a long to-do list (I dislike to do lists).

Overall, you don’t want to overcomplicate your productivity system. Something as simple as logging your daily activities can go a long way.

3 – Academia vs industry isn’t mutually exclusive

People view post PhD as either becoming an academic or going to industry. But from my experiences of both, I find a lot more people who are in both spheres. There are academics with full-blown companies on the side, likewise there are industry workers in R&D departments collaborating with universities.

It does feel like there’s a push to stay in academia after the PhD, which can be off-putting for those who want to leave. I honestly think it doesn’t have to be as binary as it looks. I love academia currently, and I would like to stay in academia. But I understand how important it is to have industry connections, whether I ever return to industry or not. As someone building multiple career paths simultaneously, I’m very comfortable with not feeling like I have to commit to one path. And to be honest, many of us doing a PhD are usually looking to pivot from an existing path, whether that was industry or undergrad.

Consider building careers that embody both elements of academia and industry. And remember that the practices that exist in academia, also exist in industry. They might look slightly different, but these two domains aren’t that far apart.

4 – You don’t have to be broke as a PhD student

There’s a common narrative that PhD students are broke. It makes sense, many universities view PhD students as cheap labour. But nowadays, stipend rates have gone up quite a bit. In most UK cities, you’re able to live well and save on the stipend as long as you budget well.

One thing I thought hard about was how doing a PhD would impact my finances. Being a CS graduate, earning big bucks was essentially guaranteed, especially given my Google internship. On a PhD you earn a lot less than you would in a standard corporate role, talk less tech. But I did the maths and realised that I would still be able to save and make money as a PhD student.

I currently have 3 main income streams: my stipend, working as a teaching assistant and creating content for the university. Most of my income comes from my stipend, but the income I make from the other 2 streams is still a reasonable chunk. It helps with saving on the stipend as well as extra spending money. With these income streams I’ve been able to save a decent amount each month, building up a 1-month emergency fund. I’m looking to build this to 3-4 months by the end of 2026.

There are multiple ways to make money as a PhD student. A lot of them work really well with the PhD structure, and some double as both experience as well as income (teaching is great for this). It’s really worth looking around and finding potential income sources to balance with the PhD. I don’t view it as an essential thing to do but I think it’s great for diversifying your income and building experience alongside or outside academia.

5 – Generative AI can help you – just use it carefully

Discussions around generative AI are quite polarising. It can feel like everyone’s trying to categorise people into pro AI or anti AI. One thing about me – I hate being placed in a box. I think generative AI is a very useful tool, but it can’t do everything and shouldn’t be heavily relied on.

I find it a great tool to assist with coding. With a CS masters, I’m already a strong coder, AI just helps me get there faster. I use it as a springboard, and I know when it’s written code that’s incorrect. However, if you’re not as strong in coding, I would be wary of using AI in this manner as you might end up getting lost in the code.

In terms of literature review and reading papers, I avoid AI. I believe that these practices should be done manually, by printing out the paper and reading it critically. Tools like Anara promising to summarise the paper aren’t helpful as you don’t retain the knowledge, you’re basically just getting a short summary of the paper. Which isn’t useful when you need to critically discuss the paper in the literature review.

The best way to think about using generative AI is to think about the utility of the task. Is the task something that benefits from being done quickly, or does it need careful reflection? I find that the coding I do on the PhD just needs to do the job, rather than be the best code ever. Using generative AI to get a starting point with a coding project can help greatly. With any task that involves dealing with literature or academic writing, using AI is quite discouraged.

Wrap-up

These are my 5 hot takes for the PhD. I’m thinking of coming up with more (once I film part 2 for the TikTok video). So, stay tuned for that!