tevon

This is true, however these known criteria can be added in, thats the whole point of the heuristic in A*.

Likely one of Uber/Lyfts greatest strengths in their algos is discovering the E[X] of those heuristics. This could probably be captured from the human dispatchers as well.

More than anything this highlights the difficulty in testing or trusting non-deterministic systems from a user perspective.

Whether or not GPT-4 is truly degraded, there will always be users who experience strange or sub-optimal responses and will be able to find other users who experience the same.

Quite a challenging space to build trust! We expect machines to act deterministically, now we as users will need to re-wire our thinking.

License plates themselves are not PII.

They are publicly available data. If the license plate is associated with OTHER PII it becomes a problem. However reading your license plate on the way into and out of a garage is just reading a publicly available code at different times of day. The fact that that code must always be displayed on your vehicle, and can be tied to your identity is an issue, but the code itself doesn't constitute PII.

This is an excellent counter-example.

Taking a strong subset of the truth may be factually correct but doesn't stand up to the standard of "the truth".

Likely because the headline implies intent (or expects its audience to assume intent), where it doesn't appear any intent exists.

We use daily as a team

You built Proxyman?! What a great project. This can't just be a side project though right?

Even if its with a small team thats impressive, the UI is polished and full of features.

Bruinen (bruinen.co) | Senior Full-Stack Engineer | HYBRID (ONSITE/REMOTE) | Full time | San Francisco

We're a small team building a better way to connect user context to AI (and other apps)—think Plaid for profiles.

https://www.notion.so/bruinen/Senior-Full-Stack-Engineer-6b9...

Bruinen (bruinen.co) | Growth Lead | HYBRID (ONSITE/REMOTE) | Full time | San Francisco

We're a small team building a better way to connect user context to AI (and other apps)—think Plaid for profiles.

The Growth Lead will define our GTM, and have broad autonomy and ownership over what experiments to run and strategies we pursue.

https://www.notion.so/bruinen/Growth-Lead-bd379fda95fe467998...

Nestjs has released a devtools site, allowing developers to traverse their backends and view changes across commits.

How big is one of these models once it is trained and tuned? Could it be run on a single reasonably-large EC2? Or does it need any special architecture?

Would love a breakdown on how this fine-tuning is accomplished. What was the format of the input data? How many prompt/response pairs was it fine-tuned on? Where/how did you extract the prompt/response pairs?

Great work btw!

The Audacity of it!

These articles really need to add a denominator to their numbers.

We currently use render. Its excellent, but is missing a couple key features: like Heroku's continuous rollback protection. If they offered this they'd be head and shoulders better.

The reputation hit to Heroku is real. Its so too bad that they've made these changes, and stopped pushing new features.

I built my prior business on heroku and it was wonderful. Now with this current company I've switched to Render (through am looking for something else). For me the writing was on the wall for Heroku and it wasn't worth it to me to dig in there if new features and support wasn't a given.

Curious for anyone with an inside perspective. What happened? Heroku was so far ahead of the pack for awhile, then suddenly stopped staying a step ahead. Was this intentional? Why?

Most people wouldn't care if it wasn't for the shady things the payment terminals do. Like reordering the tip % to either be reverse of what is expected or random.

Additionally they wouldn't care is the % was standard.

From left to right: 15%, 18%, 20%. Leave it there and stop using dark patterns.

I don't know much about lasers. Why are these phased and not continuous?

Anyone who is excels is out of their depth.

By definition if you continue to progress (in career or life), you will be doing something you have never done before. This means everyone that we look up to who appears to be on a upward trajectory is making it up as they go.

wait wait wait, the actual quote is "PREMATURE optimization is the root of all evil". Completely different meaning

I'm curious, if you're using CSS-in-JS already why not jump to tailwind? Its slightly more restrictive, but I've found the more opinionated style incredibly helpful.

What are the benefits you're getting from CSS-in-JS?

Tailwind isn't intended to be a CSS replacement (like you stated in the article).

It serves well, even in large code-bases, when you aren't doing bespoke transforms and object selection.

I've built large UIs using it and its magic. For the most part what you want is consistent spacing and reasonable defaults; Tailwind nails this.

I'd be curious if it would be better to be in the city or the country.

My hunch is that the city requires more "inputs" to stay survivable (fewer food/water stockpiles available), dependant on inputs. However cities would also be the focus for recovery efforts and disaster relief.

On the flip side, the country has a greater store of natural resources per-capita (water, likely food, heat sources), however would be deprioritized for any kind of support.

This is a very solid argument. How do we fix that?

THIS is the article I want to read!

Yes! Exactly.

The article suggests that he wants to know "who wrote the code" if a senior dev he trusts submits a PR. He doesn't want to be surprised that "the AI" wrote some of this code.

But its ALL written by the senior dev. If he trusts that dev, that means that dev has thoroughly read and tested his code! That's the important bit. Remembering proper syntax/imports/nesting levels is the tiniest piece of writing good code. And copilot can take that off our hands.

What're the examples of false positives?

Agreed it gets things wrong very frequently. But I've found it much easier to use its suggestion as another "input" to writing code.

We don't have GC (too small), so caveat my take with the fact that I'm writing from a smaller companies perspective.

May be different for a larger, value-preserving company who would face more scrutiny.

That being said, I still find it extremely unlikely that there would be legal ramifications from using a product being pushed by one of the largest software companies in the world. Why go for a user and not Microsoft themselves?

Yup. There is no chance in hell of them coming after USERS of copilot.

Seems this article completely misses the benefits of copilot. Its a massive step forward in productivity. For me, its about suggesting proper syntax across the various libraries we use. It really does cut time by 10s of percent.

I don't buy the argument that the risk of a yet-to-be-litigated case against a different company, who will certainly fight this hard; is greater than the productivity gain of using copilot.

Additionally, the security argument feels ridiculous to me. We lift code examples from gists and stackoverflow ALL THE TIME! But any good dev doesn't just paste it in and go, instead we review the code snippet to ensure its secure. Same thing with copilot, of course its going to write buggy/insecure code, but instead of going to stackoverflow for a snippet its suggested in my IDE and with my current context.

Nevermind, looks like its an on-chain add for another crypto service. Haven't seen that before

https://etherscan.io/tx/0x6bdd6c7d90c7f67223f7971745d5bc1f7b...

Text in utf-8:

Check out Fringe Finance, fully decentralised/non custodial platform. It's very very important to support projects like Fringe if we don't want events like FTX to repeat. Do check :- https://www.coingecko.com/en/coins/fringe-finance Problem won't solve just by creating Fck FTX Token. We need to support Fully Decentralised and Non Custodial platform Fringe at any cost. Lend/Borrow/Stake/LP/Stable Coin/Collaterals etc etc Team has been burning midnight oil, putting their blood and sweat into this. Check website, Read mediums. This project can end centralised scams for ever. Let's save crypto and pledge no one ever gets burnt again. Support Decentralisation, Support Hardwork, Support Fringe Finance.

Can someone deeper in the space explain the series of transactions from

"letssavecrypto.eth" to the FTX Drainer address?

Seen here: https://etherscan.io/address/0xaa16547cdc32268542145d3401120...

My guess is its a way to communicate a message via txn data.