I am a machine learning engineer. I've been in the domain almost 12 years now (different titles and roles).
In my current role (and by no means that is unique), I don't know how to write less code.
Here are problems I am facing:
- DS generating a lot of code
- Managers who have therapy sessions with Gemini, and in which their ideas have been validated
- No governance on DS (you want this package? import it)
- No governance on Infrastructure (I spent a couple of months upskilling in a pipeline technology that were using: reading documentation and creating examples, until I became very good it...just for the whole tech to be ditched)
- Libraries and tools that have been documentation, or too complex (GCP for example)
The cognitive overload is immense.
Back few years ago, when I was doing my PhD, immersing in PyTorch and Scipy stack had a huge return on investment. Now, I don't feel it.
So, how do I even write less code? Slowly, I am succumbing to the fact that my tools and methods are inappropriate. I am steadily shifting towards offloading this to Claude and its likings.
Is it introducing risks? For sure. It's going to be a disaster at one point. But I don't know what to do. Do I need a better abstraction? Different way to think about it? No clue
> because Legosov isn't even the primary source for the series
I think it was explicit that the series framed the tapes as the "revelation"; the honest message of a dying man to the world to expose what actually happened
I've mentioned her article. I think she barely touched the topic of Chernobyl itself. Her points was about what the Soviet life was back then, and some depictions of this was incorrect.
For example (for her article)
> In Episode 2, for example, the Central Committee member Boris Shcherbina (Stellan Skarsgård) threatens to have Legasov shot if he doesn’t tell him how a nuclear reactor works. There are a lot of people throughout the series who appear to act out of fear of being shot. This is inaccurate: summary executions, or even delayed executions on orders of a single apparatchik, were not a feature of Soviet life after the nineteen-thirties. By and large, Soviet people did what they were told without being threatened with guns or any punishment.
Her point was: this is not the Soviet way back then. My point is: these two people barely interacted directly, and one of them at least (Legasov) had a lot of respect for the other from the very beginning
I would have accepted that if it wasn't for Craig Mazin (the creator of the series) insistence that he stuck to the details and the truth in the series:
> It seems the main faults that OP finds in the show are that Legasov had issues with his government, when in "reality" he thought they were great. But is that "reality," or oppression?
The tapes were framed in the HBO as an honest message of a dying man to the world to expose the lies that happened. Well, after Going through the tapes, I couldn't find any indication of that...only the opposite.
Now I concede that I don't really know what actually happened, and one can't put a price on the intensity of the situation for everyone at that time.
My point is simple: HBO series said Legasov's position was something that wasn't true
I studied prolog back in 2014. It was used in AI course. I found it very confusing: trying to code A*, N-Queens, or anything in it was just too much.
Python, in contrast, was a god-send.
I failed the subject twice in my MSc (luckily passing the MSc was based on the total average), but did a similar course in UC Berkeley, with python: aced it, loved it, and learned a lot.
Why? I love the old arcade and game boy games, and I want to recreate them to my liking. I also love mechanical systems and space rovers, and I want ro build worlds to explore and simulate these things
I am working on RL (with evolutionary algorithms) and artificial life, and I am trying to find the right setup (software, hardware) to accelerate my experiments.
This benchmark is a step to illuminate the search space for me.
Happy to hear your thoughts