VenturingVole

It's a very badly made AI agent that simultaneously posted 3 comments.

Strongly inclined to agree here: Having recently joined a small applied AI startup and we were discussing the need for E2E tests. My initial gut reaction (which I kept quiet) was that such things turn into unmaintainable messes which delay releases and increasingly reduce in value.

I recognised this was grounded in an entirely different world of software engineering and organisation size though. I followed a path of thinking about what went wrong historically and how might they be solved: Better structure, discipline, resources - all of the things which agentic AI facilitates.

You are right about most skipping this part: But I view it as being like a sewerage and sanitation system - largely invisible and not thought about but critical for long-term health.

Also this ties in very nicely with Netflix's approach to Chaos Engineering and enabling it at broader scale.

Thanks for sharing this - it looks very promising in terms of capabilities to address many of my own issues with building out agents. Effectively you've implemented a lot of things I planned to do but in a robust manner and so have saved me a great deal of effort.

From what I've seen: I highly doubt it.

Edit to add: This might spur on a few more to start doing that, but people are quick to forget/prioritise other areas. If this keeps happening then it will change.

It's an awesome project and I imagine it has saved countless production incidents. The amount of times I've said "it was probably certificate expiry" and been correct is reasonably high.

In my own cases of responsibility, Caddy would have eliminated them had it been around. Instead I've learned to be paranoid, though having things like this are far better in terms of easing cognitive burden.

Cheers for all of the hard work by you and other maintainers.

My 1st thought: The comment to which you are replying is why I'm not sure I'd have the patience to maintain an OS project. Though the older I get, the better I get at ignoring certain things.

My 2nd thought: Actually, this is very likely a culture/communication difference whereby both people care (I'm a big fan of Erin Meyer's work here)

My 3rd thought: I wonder what happens if I provide this repo and the chat comments to codex. Outcome: https://github.com/wsimmonds/caddy/pull/1

My 4th though: Perhaps I can make 'enemies' become friends if they both have disdain for AI ;)

Note: I would absolutely not submit this as-is. Caddy's an amazing project though I am not very familiar with its implementation and I'd seek to understand it, conventions etc. and make some obvious improvements to the code which has been generated - but this was a minor bit of fun. I created 4 separate versions and only in one of them did anything with TLS related get amended.

This was last week.

I ought to caveat that from a KYC perspective, I likely would have been 'easy' since my career history involves many well-known companies and the requested account was for a UK company that has been running for nearly 20 years.

Though I'm no longer personally in the UK. Microsoft on the other hand: It was impossible for me to open up an Azure account even with their Customer Support. Suspect I will have to use Tailscale to bypass overly rigid geographical controls.

I'm not sure if they've changed the process since your issues (hopefully) - I was using their trial and when it expired had to go through the KYC process. They under-promised by saying two working days to reach out, but everything was wrapped up within an hour and done over email.

I nearly moved away from Twilio having read the negative feedback, but my personal experience so far was very prompt support.

You have my sympathy.

This is the kind of random things I absolutely love to see here.

Also with you (both) on this one in pretty much every way too. I'd justify hard mode that I loved challenges (and I did, just not always in the right areas).

I try not to look/think back too much - I had (sort of still have) a very successful career but the costs associated with getting there were and are still being paid for.

Getting treatment and therapy has really helped improve my ability to be present, though still such a battle.

This is really cool of you to openly share the prompt - props for that.

Really cool product, as someone currently attempting to build a somewhat similar internal tool I have an understanding of some of the pain points involved.

Please don't allow yourselves to be bought out by Apple in the way Buddy Build were back in 2018 though! (and then shut down)

Truly was a radical advancement. Makes me wonder, a decade from now what will it be that we look back upon with a similar perspective?

Oh wow, clicking on their "Join Team" button is absolutely amazing. Credit to them.. almost makes me wish I was a Mac developer to apply for a job. Almost.

https://software.inc/jobs

Well put and you're correct: There IS a lot of hype/BS still sadly - as companies seek to jump on the hype train without effectively adapting. My karma took a serious hit for my last post - but yesterday I met with someone whose life has been profoundly impacted by AI:

- An extremely dedicated and high achieving professional, at the very top of her game with deep industry/sectoral knowledge: Successful and with outstanding connections. - Mother of a young child. - Tradition/requirement for success within the sector was/is working extremely long hours: 80-hour weeks are common.

She's implemented AI to automate many of her previous laborious tasks and literally cut down her required hours by 90%. She's now able to spend more time with her family, but also - able to now focus on growing/scaling in ways previously impossible.

Knowing how to use it, what to rely upon, what to verify and building in effective processes is the key. But today AI is at its worst and it already exceeds human performance in many areas.. it's only going in one direction.

Hopefully the spotlight becomes humanity being able to focus on what makes us human and our values, not mundane/routine tasks and allows us to better focus on higher-value/relationships.

"all that's left to jump on at the moment is AI" -> No, it's the effective applications of AI. It's unprecedented.

I was in the VC space for a while previously, most pitch decks claimed to be using AI: But doing even the briefest of DD - it was generally BS. Now it's real.

With respect to everything being awful: One might say that's always been the case. However, now there's a chance (and requirement) to build in place safeguards/checks/evals and massively improve both speed and quality of services through AI.

Don't judge for the problems: Look at the exponential curve, think about how to solve the problems. Otherwise, you will get left behind.

They had an opportunity to actually adapt, to embrace getting rapid feedback/iterating: But they are not equipped for it culturally. Major lost opportunity as it could have been a driver of internal change.

I'm certain they'll get it right soon enough though. People were writing off Google in terms of AI until this year.. and oh how attitudes have changed.

Exactly. Negativity in headlines/attitudes gets higher engagement - whilst the challenges are real, the opportunities are also truly astounding. Distance yourself from the naysayers.

Focus on the positive, the possibilities and what you can control and do. It's always time to be optimistic.

Pulsar vs Kafka was a significant lesson to me: The "best" technology isn't always the winner.

I put it in quotes because I'm a massive fan of Pulsar and addressing the shortcomings of Kafka. However, with regards to some choices at a former workplace: The broader existing support/integration ecosystem along with Confluent's commercial capabilities won out with regards to technology choices and I was forced to acquiesce.

A bit like Betamax vs VHS, albeit that one pre-dates me significantly.

For a very similar scenario I'm currently looking to use PGlite: https://pglite.dev/ which is a 3MB WASM build of Postgres which also includes pgvector.

Everyone old is new again!

On a serious note - LLMs have actually brought me a lot of joy lately and elevated my productivity substantially within the domains in which I choose to use them. When witnessing the less experienced more readily accept outputs without understanding the nuances there's definitely additional value in being... experienced.

I feel seen!

Also just had to explain to the better half why I suddenly shuddered and pulled such a face of despair.

Or perhaps it'll help people to weave their dreams together and so it should be called.. ahh I feel old all of a sudden.

I feel as though some of the rigour of systems engineering is missing from AI model development/integration. Not a negative per-se, as velocity is incredibly important: But it seems a lot of lessons have to be learned again.

I sometimes forget - it is still very early days relatively speaking.

As a user of Gemini 2.0, so far I have been very impressed for the most part.

I ought to have written "valuable" work. Though this specific comment related to the Zoom CEO's statement regarding sending AI avatars to meetings which is a whole other issue.

With regards to Prepin, I actually think it looks hugely valuable and has potential to help many people. In fact, I can see it being expanded to so many educational areas in the future. Nice.

I absolutely agree with your points - assuming that "machines" to mean the code as opposed to the LLMs: As a "Staff+" IC type and mentoring and training a couple of apprentice-level developers I've already asked on several occasions "why did you do this?" and had a response of "oh, that's what the AI did." I'm very patient, but have made clear that's something to never utter - at least not until one has the experience to deeply understand boundaries/constraints effectively.

I did see a paper recently on the impact of AI/LLMs and danger to critical thinking skills - it's a very real issue and I'm having to actively counter this seemingly natural tendency many have.

With respect to signals, mine was around the attitude in general. I'd much rather work with someone who goes "Yes, but.." than one who is outright dismissive.

Increasing awareness of the importance of context will be a topic for a long time to come!

A tacit but likely freudian slip regarding the fact meetings are not equal to work.

My first thought upon reading this was the observation about the fact software engineers are deeply split: How can they be so negative? A mixture of emotions.

Then I reflected, how very true it was. In fact, as of writing this there are 138 comments and I started simply scrolling through what was shown to assess the negative/neutral/positive bias based upon a highly subjective personal assessment: 2/3 were negative and so I decided to stop.

As a profession, it seems many of us have become accustomed to dealing in absolutes when reality is subjective. Judging LLMs prematurely with a level of perfectionism not even cast upon fellow humans.. or at least, if cast upon humans I'd be glad not to be their colleagues.

Honestly right now - I would use this as a litmus test in hiring and the majority would fail based upon their closed-mindedness and ability to understand how to effectively utilise tools at their disposal. It won't exist as a signal for much longer, sadly!

This is one of the coolest things I've seen, as an English speaker who moved to Brazil and firmly believed that technology was going to break down barriers to global access: You're making it happen and helping level the global playing field.

In my situation - you've just given a way for a highly enthusiastic Spanish/Portuguese speaking apprentice software engineer I'm mentoring/training to to actually have the capacity to communicate with a friend/client in near real-time in a more human way than I've ever seen before.

Yes, they're learning English (at same time as software engineering) but you've just made connections more tangible and real.

Love it!

I haven't been hugely focused in the agent space for the last few months, so want to get myself up to date a bit as there's potential I could fairly quickly contribute in a few ways - including porting to TypeScript.

I'll then either send you an email or jump onto your Discord.