I think experiences vary. AI can work well with greenfield projects, small features, and helping solve annoying problems. I've tried using it on a large Python Django codebase and it works really well if I ask for help with a particular function AND I give it an example to model after for code consistency.
But I have also spent hours asking Claude and ChatGPT with help trying to solve several annoying Django problems and I have reached the point multiple times where they circle back and give me answers that did not previously work in the same context window. Eventually when I figure out the issue, I have fun and ask it "well does it not work as expected because the existing code chained multiple filter calls in django?" and all of a sudden the AI knows what is wrong! To be fair, there was only one sentence in the django documentation that mentions not chaining filter calls on many to many relationships.
Very good concrete examples. AI is moving very fast so it can become overwhelming, but what has held true is focusing on writing thorough prompts to get the results you want.
Senior developers have the experience to think through and plan out a new application for an AI to write. Unfortunately a lot of us are bogged down by working our day jobs, but we need to dedicate time to create our own apps with AI.
Building a personal brand is never more important, so I envision a future where dev's have a personal website with thumbnail links (like a fancy youtube thumbnail) to all the small apps they have built. Dozens of them, maybe hundreds, all with beautiful or modern UIs. The prompt they used can be the new form of blog articles. At least that's what I plan to do.
I’m not even close to being on par with other faang engineers but this is far from being a very difficult bug in my experience. The hardest bugs are the ones where the repro takes days to repro. But nonetheless the op’s tenacity is all that matters and I would trust them to solve any of the hard problems Ive faced in the past.
Hi, author here! At my job before Google I had to debug these kinds of bugs for our mobile robotics / computer vision stack, but I found them fun so they didn't feel "hard" per se. The most time-consuming one took a month on basically a camera-mounted computer vision system, where after an hour of use the system would start stuttering unusably. But the journey took us through heat throttling on 2009-era gaming laptops, esoteric windows APIs, hardware design, and ultimately distributed queuing. But fixing it was a blast! I learned a ton. I hated that project but fixing that bug was the highlight of it.
This is exactly the reason why I spend all of my non-desk time outside working out, hiking, skiing, snowboarding, archery hunting, golfing and camping. I even choose to shovel snow over using the snowblower.
I recently used AWS Textract and had good results. There are accuracy benchmarks out there, I wish I saved the links, but I recall Gemini 2.0 and Textract towards the top in terms of accuracy. I also read that an LLM could extrapolate/conjure up cropped text therefore my idea would be to combine traditional OcR with LLM to determine conflicts.
I've taken part 1 of 3 in Andrew Ng's machine learning specialization which covers the math for supervised learning, linear regression, etc. As I started part 2 (neural networks), it built off the math from part 1 such as the sigmoid activation function. This is what I think of when the OP refers to learning ML from first principles. I highly recommend Andrew Ng's course and I feel like I need to take it again to really understand those basic building blocks.
I am doing OCR on hundreds of PDFs using AWS Textract. It requires me to convert each page of the pdf to an image and then analyze the image and it works good for converting to markdown format (which requires custom code). I want to try using some vision models and compare how they do, for example Phi-3.5-vision-instruct.
Notes are extremely helpful, including those about coworkers you meet. Write down every detail during or after your meeting. Use a mind map program (miro works ok).
With the power of LLMs, notes will be even more helpful as we come up with more innovative ways to parse our daily lives.
But I have also spent hours asking Claude and ChatGPT with help trying to solve several annoying Django problems and I have reached the point multiple times where they circle back and give me answers that did not previously work in the same context window. Eventually when I figure out the issue, I have fun and ask it "well does it not work as expected because the existing code chained multiple filter calls in django?" and all of a sudden the AI knows what is wrong! To be fair, there was only one sentence in the django documentation that mentions not chaining filter calls on many to many relationships.