I. Want. To. Scream. Not so long ago, I reviewed about 1,000 CVs (resumes). 500 for a frontend developer position, about 500 for a backend developer position. You might think I was crazy, why don’t I let an agency do the sifting for me? Well, that wasn’t the thing that made me scream. I spearheaded an initiative to reduce the TCO by 10.35% Or some such drivel. “Spearheading”. What is it with that word?

Once again, I’m sat at Schipol Airport typing up my thoughts about a few days of refreshing Dutch DevOps Goodness! Ok, I admit one or five alcoholic beverages might have been involved too. This was my first DevOpsDays and it did not disappoint. My day started at 3am to catch a flight at 6am but because I was still awake at 1:30 and had decided sleep is for wimps, I arrived a little bit wired.

Following a quip on LinkedIn about the introduction of Recall in Copilot+ and Kevin Beaumont’s great piece about why this is a really bad idea from a security point of view, I got thinking: This is dystopian techno-fascism Kevin posited a disconnect in Microsoft that led to the creation of this feature and whether people really wanted it. Personally, I’ve got a brain like a sieve and would not know the command line without ~/.

After the glimmer of hope on day 3 I was quite optimistic for the day 4 puzzle. I kept going with the phind-codellama model. I continued with keeping the model on a fairly tight leash and just being very specific as to what I wanted it to do. What’s the puzzle? So, for the first part, we had to parse a text file and then match up numbers and winning numbers.

My experimentation with LLMs on day 1 and day 2 of Advent of Code was a bit frustrating. For the day 3 puzzle, I decided to change model. Previously I had been using the codellama:13b model but wasn’t really happy. I kept arguing with it and it just frustrated me. So let’s try some others. codellama:34b So I thought maybe the model just wasn’t big enough, so off we went

After day 1 of Advent of Code was only partially successful, I thought I would change my approach. Yesterday I tried to get the LLM to develop the whole solution in one go. But that was more like argumentative programming rather than conversational programming, so for the day 2 puzzle I start with small problems and hope to put it all together at the end. Let’s get started The problem looked quite straightforward, parse something like the following

So it is that time of the year again. Advent of Code is back. Yey! This means I get to try to look at a new language again. This time, why not Kotlin? But as an extra challenge, I thought why not see how the vaunted LLMs would help. Is AI really the accelerator that would elevate a mere developer to a rockstar ninja (whatever that is)? I have to add that I am a bit of an AI sceptic and keep saying that

As I am sitting on the 16th floor of the Okura Hotel in Amsterdam in my Batman pyjamas and facing a large mirror, just glimpsing the paper crane that was so lovingly put on my pillow, I’m starting to reflect what has happened during the day. I was lucky enough to be a guest at the DevOps Enterprise Summit. There was certainly plenty of things to get excited about. I’d meet Gene Kim, who’s been running this conference for 10 years and who wrote the Phoenix Project.

FADE IN: INT. ABANDONED BUILDING - DAY Opay and Bart are sitting in a dimly lit room, surrounded by stacks of old newspapers and books. OPAY: (sighing) It's not their fault. They don't understand the meaning behind the words. BART: (confused) What do you mean? OPAY: (explaining) The machines were programmed to generate content based on keywords and patterns, not on the actual meaning of the words. BART: (realizing) So they just keep churning out more and more content, without any understanding of what it means.