I personally review most of the code written by my team without using an LLM.
Just today I reviewed a patch made by a teammate using Claude for a not very significant bug, where in addition to fixing it (in a very roundabout way), it introduced 2 new very subtle bugs, and refactored a large portion of another function that was unrelated to the code, and made several observations and comments which were subtly wrong and out of touch with the rest of the code.
I ended up rejecting the PR and writing my own very small local patch for the bug.
Other people in the team use Claude to help them review my own PRs, and it sometimes finds minor issues, although there was only a single time it found a major issue.
My own use of LLMs is mostly to develop large features, which I then thoroughly review myself.
Funny, I'm just reading War and Peace myself (the Anthony Briggs translation) and having the same reaction, gushing occasionally to people I know how approachable it is, and how darkly funny and modern it feels. Well, at least after passing through the first ~200 pages which are a slog. I didn't find even Tolstoy's historical musings boring, although he tends to repeat himself. And I usually suck at names, but the main characters are done so well I find them easy to remember. There aren't that many important ones despite how it seems at the start. It also serves as a fascinating peek into the daily lives of Russians of all stripes in the early 1800s.
I also had the same reaction to Crime and Punishment as the OP did.
Read those first 200 pages 10x could never get past it. 300 characters with names that I’ll never remember, some woman and her son, a general or something. A guy that keeps saying “Capital!”, standing around at parties.
I’m sure it’s good but I don’t think I have it in me to try again.
I swear it took me six retries to make it past the start. But if you have six hours the BBC adaptation is pretty good IMHO and captures many of the essentials of the book if not all the details. The show made me cry and the book did not have the same effect but maybe that was because it focused on certain aspects. I particularly remember the combat scenes in the book would have been difficult to match - the prose capturing the chaos and randomness of brutality in the neighborhood of D Day landing in Saving Private Ryan but with horse cavalry charges and cannon fire.
I listened to the audio book version of War and Peace. I think it was something like 25-30 hours. The audio format helped keep the pace going and also it helped with the names. Although for some things, the audio format made it harder to look up in the dictionary, like I kept hearing agitant instead of adjutant, so that part didn't make sense in a lot of the military scenes. I agree with the parent that the book was very engaging, parts even felt like I was watching a movie, e.g. the drunken party tying a bear to a police officer, the foxhunt scene, the duel, the battles like when Petya gets shot, and the burning of Moscow. I even liked the abstract ending when Tolstoy relates human history to calculus so that each individual person has an infinitesimal but real contribution to history.
Ha, I'm a gardener then, on my 15th year of maintenance. So halfway there according to you. Slowly, very slowly, fixing the thousands of bugs the rockstar left behind 15 years ago.
It's easy to look at a newly made garden, its flowers all freshly planted and its paths all neatly laid out, and think it finished, but once you see one that's been carefully maintained for decades and how its plants have matured and grown together and how the steps have been worn by human feet, it's clear the project is merely off to a good start.
This comment made me imagine what would the guy maintaining my first deployed project at my first company thinks about me. Never had this thought ever cross my mind.
I work on an ancient codebase, C# and C++ code spanning over 3 major repos and 5 other minor ones. I'm senior engineer and tech lead of my team, but I also do a lot of actual coding and code reviews. It's a somewhat critical internal infra.
I'm intimately familiar with most of the code.
I've become somewhat addicted to using coding agents, in the sense I've felt I can finally realize a lot of fantasies about code cleanup and modernization I've had during the decade, and also fulfill user requests, without spending a lot of time writing code and debugging. During the last few months I've been spending my weekends prompting and learning the ropes. I've been using GPT 5.x and GPT 4 before that.
I've tried both giving it big cleanup tasks, and big design tasks. It was ok but mentally very exhausting, especially as it tends to stick to my original prompt which included a lot of known unknowns, even after I told it I've settled on a design decision, and then I have to go over its generated code line-by-line and verify that earlier decisions I had already rejected aren't slipping into the code again. In some instances I've had to tell it again and again that the code it's working on is greenfield and no backwards compatibility should be kept. In other instances I had to tell it that it shouldn't touch public API.
Also, a lot of things which I take for granted aren't done, such as writing detailed comments above each piece of code that is due to a design constraint or an obscure legacy reason. Even though I explicitly prompt it to do so.
Hand-holding it is a chore. It's like coaching a junior dev. This is on top of me having 4 actual real-life junior devs sending me PRs to review each week. It's mentally exhausting. At least I know it won't take offense when I'm belittling its overly complicated code and bad design decision (which I NEVER do when reviewing PRs for the actual junior devs, so in this sense I get something to throw my aggression against).
I have tried using it to make 3 big tasks in the last 5 months. I have shelved the first one (modernizing an ancient codebase written more than 20 years ago), as it still doesn't work even after spending ~week on it, and I can't spare any more time. The second one (getting another huge C# codebase to stop rebuilding the world on every compilation) seemed promising and in fact did work, but I ended up shelving it after discovering its solution broke auto-complete in Visual Studio. A MS bug, but still.
The 3rd big task is actually a user-facing one, involving a new file format, a managed reader and a backend writer. I gave it a more-or-less detailed design document. It went pretty ok, especially after I've made the jump to GPT 5.2 and now 5.4. Both of them still tended to hallunicate too much when the code size passed a certain threshold.
I don't use it for bug fixing or small features, since it requires a lot of explaining, and not worth it. Our system has a ton of legacy requirement and backwards compatibility guarantees that would take many days to specify properly.
I've become disillusioned last week. It's all for the best. Now that my addiction has lessened maybe I can have my weekends back.
I've been using Perplexity for small, fast queries almost exclusively for the last year or so. Their Sonar model is Llama running on top of a Cerebras chip, and searches the internet in an incredible speed. Its results are astonishingly good (for a Llama model), although in more niche areas it still makes mistakes, so in those areas I usually double-check its sources or do an extra ddg search myself.
Actually I've never used chat gpt, I went straight to Perplexity after having discovered it. Their free tier is extremely generous (not even requiring an account). Not affiliated.
OP currently doesn't look it will affect that, seems like Open AI touts it for agentic coding only, not as an alternative to chat gpt, although that will probably change.
Just today I reviewed a patch made by a teammate using Claude for a not very significant bug, where in addition to fixing it (in a very roundabout way), it introduced 2 new very subtle bugs, and refactored a large portion of another function that was unrelated to the code, and made several observations and comments which were subtly wrong and out of touch with the rest of the code.
I ended up rejecting the PR and writing my own very small local patch for the bug.
Other people in the team use Claude to help them review my own PRs, and it sometimes finds minor issues, although there was only a single time it found a major issue.
My own use of LLMs is mostly to develop large features, which I then thoroughly review myself.