The Value of Time and Struggle

I have many conflicting thoughts that I - as a data engineer, a consultant, a member of society, a friend to some, and husband to one - am going to try to articulate throughout the course of this post. I will do my best to organise my thoughts throughout this piece in way that at least helps me formalise how I am currently feeling about my long-term future in work due to the present experiences I have lived over the last two years as AI...

Not that word!

Let's get one thing out of the way. I can hear you all thinking. Surely he doesn't mean LLMs mean AI, they're not intelligent! We probably agree that they aren't. We could also probably have discussions about whether we as humans are doing similar things that models are, just with hundreds of millions of years of evolution behind us. Intelligent or not it's a term that's universally understood to be a catch all for this new swath of technologies grounded in incredibly impressive statistical models that are able to emulate human behavior in ways that is in many ways shocking, upsetting, and impressive all at the same time.

I, like you, may take issue with the fact that many of our collective efforts saw no compensation in respect to the creation of these model's training sets. Books, articles, songs, art, it's endless and I recognise all of that. It's upsetting, but it has happened. That is not the point of this post. For the sake of brevity for the rest of the post, AI means the above.

Who is this guy?

My background is in 10 years of programming and data engineering after two business degrees in finance and international business. My understanding of large language models from a theoretical perspective is limited, but I have extensive use in using them across a variety of use cases and have been following the progressions of such for years, as it is incredibly important to stay on top of because of my job. I am fortunately in a position of influence where I feel a duty to not only work out for myself how my industry is evolving, but for others I work with and sometimes lead or influence. Therefore, I feel like I have at least enough experience and responsibility to talk about it at a high level.

Just in the same way I rely on doctors to give me the best treatment and I trust their judgment and inferences. I think it's not controversial to draw conclusions from experts in the industry of what these technologies -- when boiled down to their purest form -- are.

Finally, I'm not going to guard all of my positions with the opposing view or say things like, "I, like many of you may, feel" all of the time to make it sound like I'm special. I hope you hear in the sincerity of my tone, that I recognize that I am not a distinct case, my experience or importance is nothing greater than what you should feel about yourself. We've all worked very hard to get to where we are in life, and I'm going to be talking about it from my perspective. I don't want to invalidate the effort and feelings that you have put in. It just takes a lot of time and effort to guard every sentence in that way and it's not very fun to read.

Okay.

Am I Special?

For much of my career and earlier life, I felt that there was a direct correlation between the effort, performance, and outcomes that I put into things helped separate me from somebody who had done something less efficiently, beautifully, or valuably. They were distinguishing factors about my work and my colleagues' work that helped provide certain pieces of value that at least helped my firm stay competitive, win work, and differentiate us from other consultancies' approaches or technical abilities.

It felt like I was on a great basketball team with a great group of guys that allowed for us to make it to the top of the league. Whereas now, it feels like the time and effort it took to develop that team was for waste; as you can simply enter your A+B+A+B+L+R+L+R cheat code to get all your players topped out on the technical skills that you need to be competitive. Whilst at the same time, every other team in the league found the same cheat code and now we have the '92 dream team playing against itself... on repeat... for what looks like the foreseeable future.

In some ways, it's nice to think that I wouldn't need to go to the gym, practice dribbling or taking charges for years to get to the point where I am now. It's hard work that requires determination, sweat and sometimes blood. To be able to just go into that chamber and come out as Captain America or Goku is a tantalizing prospect until you realize that every other person can do the same thing.

Where do I compete now? I must compete on the softer side of business where relationships are forged, conversations are had, and decisions are made. Luckily I've had a tremendous amount of luck in meeting incredible people over the course of my career and that makes the prospect of doing this as a differentiator on existing projects or clients not so bad. Being flexible, friendly, and helpful are all things that I love to do anyways for people I work with because it helps to make their days a bit better. It's valuable to the teams or projects I work in, helps in driving motivation and perseverence to the goals we set for ourselves.

But how am I able to articulate that value when they're mostly hiring me for something else deemed to be, and almostl certainly is, much more technically complex, requiring years of experience and education? Working independently, it would be easier to make that case, but as part of a larger business where I find myself (and where many even larger consultancies find themselves) is that it's more challenging to use individual personas or people as the main USP to win multi-million £ contracts. That factors in sometimes but mostly we are going to be hired for our technical skill. But in the current landscape I worry that our differentiators are something that we can't bake into the presentation or bid as easily, because it's now much easier for the stragglers to look like the leaders.

When everyone has the dream team at their disposal, none of the technical skill matters nearly as much.

To some people, that may not matter much because they may derive value from seeing something delivered and take no pleasure in the process of actually building it. They may find the softer side, the decision-making, the most satisfying part of the job. I know I quite like that part of the job, but I would be lying if I didn't feel nostalgic for being able to be able to sit down for a few days and be able to fully justify spending the time to tinker and play with different solutions.

To learn, to grow, to feel full with the knowledge that I've gleaned over the course of the day. This point may be my own limitation. If I had a task that would take three hours in the past, I would feel satisfied. If I managed to do it in 30 minutes now with the help of AI, I will find more work to fill that two and a half hours.

Some will say, "Oh, well, why don't you take that earned time of productivity back for to yourself". It's that it doesn't feel earned to me. It feels empty. I don't learn as easily when I don't do it myself. It's easier to review something and play with something that's been given to you than it is to write it from scratch. I want to feel satisfied.

Some of that is a bit over the top, I recognize. Being able to utilize these models has helped me learn by being able to have conversations in a way that makes sense to me; to explain concepts that a textbook may not be hitting just quite right for me. Equally, I still feel immense satisfaction when I see the faces of others after I've helped them realise something that they've been trying to do for a long time, receive good feedback for my work, or help make someone's day a bit better by helping. I'm just calling out how it can feel sometimes because it's important for me to see it written down that it's okay to take a break.

Perhaps I need to work more learning into my schedule. I feel strongly about using AI to only generate things that I can fully understand and could have written. It helps me mentally anchor the output to something that I have made. If it is not, it becomes a magical black box that just works and that is both deeply unsatisfying and irresponsible. It's still me who is responsible for the product, outcomes and design. When things break, pulling the slot machine to fix it can make things worse, drifting ever further from the answer. And that is why the learning is so crucial. Whilst I can offload certain tasks to AI, I should be using that time to continue my understanding in other areas that are new to me, so that my responsible use of AI can expand and expand to more and more things that I understand.

What's It All For?

The most interesting side effect to me when working across a multitude of problems at once, is that my desire of learning something for the sake of learning something has undoubtedly been diminished. I had a great time reading a book about learning Elixir and the Phoenix framework and had an enjoyable experience building a couple of websites to utilize live views. I could spend weekends happily programming for fun, learning about different concepts and getting better. Every now and then I still feel that inspiration and will dive into things, but it's undoubtedly become more rare after AI's capability explosion. It has gotten to the point where it's hard to sometimes justify the effort or the point in learning something like that. That reason being able to call that time valuable.

Previously, I knew that my more niche programming learnings would be something that could be a differentiator alongside my personality skill set and value that I can provide to others. As someone who is a self-taught data engineer, this was deeply important in helping to eradicate much of impostor syndrome I felt. I feel incredibly fortunate that experience happened between 2015-2020 instead of 2025-2030. Now, that skills potency has been diluted greatly due to the effect that people only need a cursory knowledge of something's existence in order to get to where it may have taken me months to get to. Not in terms of understanding, but solutions built off of similar things.

This is probably a topic for another day, but it does make me worry for the motivation of open-source contributors and developers of more niche languages or solutions that are not being suggested by large language models. If someone wants an app nowadays the Next.js React full stack app is almost ubiquitous at this point and I feel for the long-term motivation of resisting when there's so much the world has to offer in terms of experiences that don't feel like fighting against the amoeba of AI generated solutions. That said, value and satisfaction comes in all kinds of places and it's not my place to say what somebody should spend their time on.

A positive side effect has been that it has shifted my priorities back to my home, community and trying to find places where I can still exist as an individual and be seen as distinct from others. This mostly just means I go out on more walks with my partner and dog instead of learning a niche throwaway skill that could have been useful to know five years ago, so I'm not really complaining. I'll just try to fit learning those niche skills during working hours whilst the bread and butter tasks are being worked on by agents.

So Where Am I Going With This?

My general takeaway and the crux of my revelation is that businesses are going to have to work out whether their good or service -- and the people that they employ to deliver that good or service -- values the struggle and the time sacrifice that individuals put into their job in order to create that product. For example, I think many people would agree that a t-shirt from a high street shop that's worth five pounds and was mass-produced doesn't hold a tremendous amount of value because the labor costs that go into producing it are negligible due to automation. This is in stark contrast to people much more willing to spend hundreds of pounds on a custom handmade pair of shoes. However, if that shoemonger decided to start mass-producing that pair of shoes, despite them being identical to the original pair that took weeks to make, they would almost certinaly be perceived as less valuable and the cost would be less to buy. Great for the consumer though.

This can also be seen as better for the producer in this example as they can sell more shoes, but the analogy doesn't really apply to the knowledge based industries as much. If they were analogous I suppose it would probably manifest in an explosion of more knowledge based services and solutions being explored for lower costs to deliver. Multi-year projects condensed to half-year projects would mean that we'd need multiple smaller projects to fill the gap of productivity and influx of money for delivering it. (It's not like it'd be the same price to deliver the same thing in a quarter of the time). This might be possible in the short-term but I -- perhaps ignorantly -- am failing to see how scarcity and climate change doesn't eventually come to bite us. We already have no shortage of things to do with our days, how much more can there be without exhausting it all? I guess we'll need all kinds of solutions to the problems we've built for ourselves.

There is a real backlash against AI artists on Spotify and I've yet to see an AI generated painting in an art gallery over the last three years. I don't think that it's because a machine couldn't eventually generate something of a quality emulating humans. It's the fact that the AI didn't need to sacrifice anything in order to create it. And the time and effort that humans have put into teaching themselves the skills to give back to society to earn a living are often very correlated with the value of a particular good or service.

Of the many software engineers around the world that are dramatically more brilliant in many ways to myself are almost certainly facing similar revelations that all of that effort will simply serve to proofread output and volunteer them to be those holding the responsibility of a product they no longer need to build. Thankfully, I think many of the technical skills are so specialized, including my own, that it is still very much necessary to check and validate the output of these AI models line by line to ensure that it's something that will actually be architecturally sound, will work well, and meet all expectations and edge cases. I don't necessarily think that all of our jobs are going to go (Just look at how Github's number of monthly commits has exploded over the last year), but I do have a sneaking suspicion that they will become seen as less "valuable".

Less Valuable?

There need to be fewer people doing the doing, and we all can't be decision makers with clients, or we might be facing a too many cooks situation. (As a side note, I have noticed in my own experience that often times the code is not the bottleneck, it is these decision points and design decisions that just can't seem to get everyone to sign off on them). Because there are fewer people doing the doing, there's more competition and more desperation to get into one of these jobs. Over time, I could see a flat-lining or depreciation of software engineering salary due to over-saturation of supply of engineers to the market.

Secondly, (and perhaps this can be seen as a lifeline depending on how you look at it) ideas themselves will become less valuable. Ripping off other people's solutions is dramatically easier now than it used to be. This may cause a similar effect to how there is a real ceiling to what you might pay for something when there's a thousand imitations all doing similar things.

And thirdly, as you might suspect I'll say, there is a less perceived sacrifice of a person's time that goes into building certain products now. No matter how technically difficult it was for a model to build it, if it can build it in a day, there is a ceiling to what someone is willing to pay for it because they can just try to vibe it themselves if they don't like the price.

Let's Wrap It Up

I feel like this has already served as enough of a therapy piece for me that I feel that I can wrap it up. There have been some real benefits of the emergence of AI in my work that has led to me being able to learn more quickly. At the same time, whilst I can deliver more work, sometimes my patience and desire to work quickly and context switch back to something else means that I skim read more than I used to. This had an unintended side effect in me getting back into reading so that I could practice staying focused and improve my ability to read for hours as a mechanism for me to stay focused on particular tasks at work. So in that respect, it's been positive, as I had forgotten how enjoyable it was, and how useful it is to maintain my ability to read long-form content.

Democratising knowledge also has given me back evenings of upskilling and has also given me the ability to compete with others who have more knowledge than I as well. It is just a change to my life that I am still adjusting to and whilst this post may have come across as overly negative, it's undoubtedly had some real positive effects as well, no matter how temporary they may be.

What most inspired me to write this was me just trying to articulate why I feel excited and nervous for what my life will look like in ten years. I suspect I'll either be doing the same thing because AI growth flat-lines still isn't good enough to do my job fully from a technical perspective. But if we progress at the rate that we have been, I could see the economic and profession impacts drive me (and perhaps us all) in a race towards some form of entrepreneurship where we all remind ourselves what is valuable in life, and try to convince each other why they should spend their UBI cheques with me.

Thank you for reading, I hope you have a great rest of your day and feel free to reach out and let me know if you have any thoughts that may help broaden my perspective on things as well. I'm always happy to grow and learn new things. If you need help with any software, web, infrastructure, data problems as well, please feel free to reach out. I'm sure I'll be able to help in some way!