iot_devs

> Originally this was caused by an Istio sidecar pod reaching its concurrency limits and failing to auto scale correctly because of a misconfigured policy that watched host service but not sidecar limits.

I operated services at similar scale, and generally we use to put a bit of slack so that you would get an alarm when capacity goes up to 80%+ (or whatever number makes sense)

This allows to check, in the morning, after coffee, why the load balancer fleet didn't scale up automatically.

I am sure there is a good answer to why this is impractical, but it would be nice to know

I personally read the thinking traces to know if the model is on the right direction

I put it online and accessible for people to try! Any feedback is welcome!

https://structo.redbeardlab.com/

A writing app.

The overall idea is to work/write at an higher level of abstraction where you work with "ideas" (just small paragraphs).

You can move them around and basically restructure many/whole paragraphs by simply changing few words.

The final text is generated by an LLM.

LLM that has the whole context so you can ask it questions, but more importantly it can provide you feedback

How is ironcalc different from other spreadsheets?

Absolutely!

Also if this scale anywhere close to LLMs in 2 years they will fold at super human speed and much better than me.

The caption for the WAS aware clearly mention 52 US states.

Is that a typo or some quirk of radio amateur?

I am afraid that this post is missing the biggest point.

Given a prompt (or task) how do I evaluate if it is a "simple" task that should be executed by a small model or if it is a complex one that may need a SOTA model?

Are you guys using heuristics? Which one?

Do you have any example?

I would love if you could make some examples

Why I believe to be different this time is that the engineer themselves feel this need AND this is a real need making software delivery actually faster.

We will see!

What did you learn so far?

I am interested in a similar tool and it would be nice to skip some of the learning

> Also the majority of people just don’t care enough.

Yet! It is the next frontier and we will need it for having agent as described in the post to really work

How descripting is the caption that you obtain?

So you include colour, shapes, etc?

I mean it is hard to argue with the number - I believe everyone will benefit for more interconnections. More energy to Europe, more money to Spain.

And for the numbers it seems obvious that renewables are a fundamental part of the picture.

Maybe it is me, but I am much more interested in a way to maintaining a wiki.

Ideally something that doesn't require rescanning all the sources and something that can verify that the wiki is correct.

Something that I never quite understood is who lend the money for this sort of activities?

The lender knows how risky they are.

I thought it was named after the characters of Suits: Harvey and Mike

Yeah, but my projects are personal and not popular.

I crafted the AI loop to do exactly what I would be doing by manually.

Out of 10 PRs, 6 to 7 gets merged. The other simply get closed.

> Give an agent the right interfaces and it becomes less conversational and more ambient. It no longer needs to constantly ask, explain, summarize, and negotiate. It can stay in the background, react to changes, and make steady progress with less supervision and less noise. That is closer to Weiser’s vision: calm technology, but for machines.

I tend to agree quite a bit.

I created a ambient background agent for my projects that does just that.

It is there, in the background, constantly analysing my code and opening PRs to make it better.

The hard part is finding a definition of "better" and for now it is whatever makes the longer and type checker happy.

But overall it is a pleasure to use.

On the other hand, I found MiniMax M2.7 a reasonable model that I could trust.

I guess really depends on tastes

Someone can explain to me what's the expectations for these AI labs?

I mostly see their products as commodity at this point, with strong open source contenders.

Eventually it will become hard to justify the premium on these models.

Honestly I don't see the problem.

The students are cheating into studying more?

Homework and home assignments are not really a way to grade students. It is mostly a way to force them to go through the materials by themselves and check their own understanding. If they do the exercises twice all the better.

(Also nowadays homework are almost all perfect scores)

Which is why LLM are so deleterious to students. They are basically robbing them of the thing that actually has value for them. Recalling information, re-elaborating those information, and apply new mental models.

Are educators reading this posts?

My SO is a college educator facing the same issues - basically correcting ChatGPT essays and homework. Which is, beside, pointless also slow and expensive.

We put together some tooling to avoid the problem altogether - basically making the homework/assignment BEING the ChatGPT conversation.

In this way the teacher can simply "correct"/"verify" what mental model the student used to reach to a conclusion/solution.

With a grading that goes from zero point for "It basically copied the problem to another LLM, got a response, and copied back in our chat" to full points for "the student tried different routes - re-elaborate concepts, asked clarifying question, and finally expressed the correct mental model around the problem.

I would love to chat with more educators and see how this can be expanded and tested.

For moderately small classes I am happy to shoulder the pricing of the API.

I find that this approach matches my experience but it is difficult to get junior engineers to stop using LLMs without critical thinking.

Justin, as a friendly suggestion, double check your recruitment process.

I believe you guys are rejecting people based on keywords instead of reading CV.

I applied to a position and I believe to be a great fit, but in my CV you won't find the keywords that appears in the job description. Automatic rejection in no time.

Nothing wrong, but you are leaving a lot of talent to the table, and I was very hesitant in applying giving all the comments I see in your posts.

Got it and reply!

I must make the email in the homepage more prominent

I definitely don't mind!

Feel free to use the email address on the GabrielAI website!

I'd be happy to chat about it!

I am working on GabrielAI, which is a tool to filter and autodraft reply for Gmail and outlook and of course it uses LLM under the hood.

https://getgabrielai.com

In cases like this the author could just set up a filter like:

"If the email contains a task for me" (or some variation)

Then add a Gmail label to it.

In this way the author will immediately find all the actionable emails for him in a specific folder, much faster to skim and to keep track of all of them.

Another option would it be to have GabrielAI generate a Draft like "reply acknowledging the task and put a to-do date in the email in 1 week"

This would allow Google to track the email and the deadline.

I loved the example with Gmail because I am the main hacker behind GabrielAI.

Basically smart filters for gmails and Outlook.

You got to express in plain English which email to filter on and how to create a DRAFT and the tool automatically filters your email and generates drafts or add labels.

It was born out of frustration of replying to several emails, all the same, with content that was already available online. Still need to provide a human and technical-ish touch.

Demo video on: https://m.youtube.com/watch?v=ALQNDYjLQUU

https://getgabrielai.com