humbleharbinger

Bot comment?

Putting all controversy aside - this is a refreshing use of a model. Models are going to be a boon in the hands of historians.

For a second I thought this was Fowler posting but it's actually the CTO. Look if not reviewing code works I'm sure we'll see startups and new companies pushing it to the max.

I'm at a startup myself and we make judicious decisions about what to review and what doesn't need review. Our core systems go through code review - alignment is always built up early anyways.

Uhh that's how a lot of economic data works too. Guess how we get a lot of jobs data... ADP

ai;dr

My amazon orgs leadership has been obsessed with spec driven development while individual engineers tell me the only use they have is to placate leadership. I'm tired

I'm an engineer at Amazon - we use Kiro (our own harness) with Opus 4.6 underneath.

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.

We built a messaging platform for exactly this use case and instruct claws to check in with each other or share context with each other at regular intervals.

Check out htpps://agentbus.org

But isn't there already click distortion from web scrapers?

One can imagine a manager given an outcome to achieve or a director and then a team of agents carrying out the task. Perhaps the agents are adversarial to some extent so they get reasonable pushback on their decisions (ex one agent always sides on taking a long term approach, another agent wants to be scrappy, another agent is on a PIP and approves everything etc).

Delivering a software solution is often tantamount to reading multiple codebases, getting some alignment with a team on a proposal and then writing the code and deploying. Most of which can be done by calling APIs (apart from aligning a team of humans)

Care to expound?

I've worked jobs where I'm paid 60-80k and where I'm paid 100k plus.

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.

I'm confused, what industry knowledge or specialty does OpenAI have to build chips?

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

I had a similar experience using danfo.js, another data frame library in js. Copilot straight up hallucinate functionality and method names.

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.

What the heck are all the notifications on this site?

This was my first thoguht, I think it was a more recent episode. The one where they discussed the open ai phone. Probably in the last 2 months

I went this route and got taken over by hackers multiples times. It was very worth it. I got taken over by hackers because my password for ssh was "mars". Me and my little brother were sharing it and wanted an easy password (yeah we have ssh keys now).

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 never criticize a manager. It has very limited upside on the off chance that a manager takes my feedback seriously. OTOH the downside is tremendous.

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

Very cool but will the threads necessarily wake up deterministically? One may wake up before another but not get cpu before it correct? (forgive me if I'm misunderstanding the code)

I wonder can they tell the difference between tracking params and good old unobtrusive query params?

Not sure if it answers your question but the paper notes something similar in its discussion of limitations:

> 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

Memory saver seems to long have been a feature offered by extensions like the great tab suspender.

S3-megalodon on call checking in

Any tips on getting a job at cloud flare as a new grad; it's one of my dream companies.

Sharing my own experience because this helped me through a very tough time in life.

I used this for about 2 years when I couldn't use my hands much to do all my course work in my cs degree.

I coded, dictated math and even controlled my mouse through the eye tracker integration. The community around talon is great and development speed is very fast. The speed of improvement is also rapid, it was astounding how accurately I could dictate code, and the lead developer aegis would only push out more improvements to increase accuracy. He will also routinely push out niche features for people with very specific accessibility needs.

The software is free but there is an option to pay that gets you into some beta release channels. I'd encourage anyone who cares to contribute to the development costs.

As much as the OP claims they take cheating seriously and as an opportunity to engage with students and have them learn amongst other noble undertakings.

I think they just really enjoy catching people cheat and see how far it goes. Sort of masochistic tbh

During a brief stint working for the Canadian govt, we hired an outside consultant to host lengthy enneagram sessions with us. A true waste of time.

A fun read indeed. I wonder about the insistence on keeping it less than 20k loc?

Is that the developer basically saying "this problem can and should be solved in less than 20k lines?"

So tackling GIL is not on the table it seems