Most of my gripes are with the harness, CC is way better.

In terms of productivity I'm def 2-4X more productive at work, >10x more productive on my side business. I used to work overtime to deliver my features. Now I work 9-5 and am job hunting on the side while delivering relatively more features.

I think a lot of people are missing that AI is not just good for writing code. It's good for data analysis and all sorts of other tasks like debugging and deploying. I regularly use it to manage deployment loops (ex. make a code change and then deploy the changes to gamma and verify they work by making a sample request and verifying output from cloudwatch logs etc). I have built features in 2 weeks that would take me a month just because I'd have to learn some nitty technical details that I'd never use again in my life.

For data analysis I have an internal glue catalog, I can just tell it to query data and write a script that analyzes X for me.

AI and agents particularly have been a huge boon for me. I'm really scared about automation but also it doesn't make sense to me that SWE would be automated first before other careers since SWE itself is necessary to automate others. I think there are some fundamental limitations on LLMs (without understanding the details too much), but whatever level of intelligence we've currently unlocked is fundamentally going to change the world and is already changing how SWE looks.

Check out htpps://agentbus.org

Analysis is just calling APIs to get data. We use that to drive a decision and then we write code and use APIs to deploy it. We close the loops with the same data/metric APIs to verify our code achieved an outcome.

I wonder if this is why SWE seems so susceptible to automation - LLMs are great at chaining together calls to APIs.

Just a thought - I know there's a lot of diversity in SWE and this may not apply everywhere. I can think of holes in this line of reasoning (ex. we make decisions about the product during this process) but want to hear others' opinions.

Comments URL: https://news.ycombinator.com/item?id=47284463

Points: 1

# Comments: 10

I can tell you for sure I cared much less about the job where I wasn't getting paid as much. I would never answer an after work call or go out of my way to follow up organizational or systemic issues. For me working overtime is part of the package when I get paid loads more.

They're a software company that works higher on the stack unless working on cutting AI has given them an edge on chip design.

Not a big deal because I just read the docs but it was annoying that I couldn't have copilot just spit out what I need.

Anyways, we both learnt a lot (htop tmux etc) . I'm always jealous that he got to learn everything earlier than me. But if he's not better than me then I consider myself a failure wrt being an older brother.

The only drawback is that this doesn't work if you want to do ai stuff. For those use cases I rent a machine on paper space for a cheap hourly rate.

I don't think feedback in a corporate setting from someone you have power over can be relied on.

> the linear attention of RWKV leads to significant efficiency gains but still, it may also limit the model’s performance on tasks that require recalling minutiae information over very long contexts. This is due to the funneling of information through a single vector representation over many time steps, compared with the full information maintained by the quadratic attention of standard Transformers. In other words, the model’s recurrent architecture inherently limits its ability to “look back” at previous tokens, as opposed to traditional self-attention mechanisms. While learned time decay helps prevent the loss of information, it is mechanistically limited compared to full self- attention.

> Another limitation of this work is the increased importance of prompt engineering in comparison to standard Transformer models. The linear attention mechanism used in RWKV limits the information from the prompt that will be carried over to the model’s continuation. As a result, carefully designed prompts may be even more crucial for the model to perform well on tasks