I had a personal development account for ~15 years. I tinker with infrastructure stuff and had built some centralized event reporting. One day about two years later I turned on sqs data events into cloudtrail. What I didn’t realize was that this closed a feedback loop and over the next couple of hours my run rate went to about $4k per day in cloudtrail+sqs usage.

I didn’t realize it until I hit the next months billing alarm immediately the next month. I’d racked up $25k in usage fees.

I’ll be using this feature. Nothing I run is worth that risk.

Which will likely get cut off at some point.

Revolutionary? With an appropriate harness could could do the same thing with the vast majority of modern large language models.

It's just so much faster and so much cheaper that it feels qualitatively different.

I also think it represents a bit of validation for folks looking for ways to bake models into hardware. Sometimes it's perfectly appropriate to sacrifice good for fast and cheap. I haven't asked Jev to do anything that GPT-3.5 would have likely done worse with.

The fly brain thing on the other hand.

Fed Claude an api key from typesafe and a link to documentation, and within about 10 minutes I had a view of HN that was populated with a little ranking as to sloppiness of each comment.

When your mind has been wired a bit to LLM latency, it feels extremely fast, and for such a subjective rating I think it did a good job.

Feels like it sits in a space between traditional ML classification and the frontier models. I can't think of a 'real' production use case for it in my sphere of influence, but certainly some will. And of course there will be five Jev competitors by the end of the year.

3 kinds of users that don't scare the shit out of you.

My buddy works with lots of folks that have gone all in on Harvey. There are other kids of users.

Writeup was very refreshing to read, felt like someone actually wrote it.

Comments URL: https://news.ycombinator.com/item?id=49715335

Points: 1

# Comments: 1

Focus on the opportunities rather than the problems, connect people doing similar things in different areas and jump in when they are solving problems after the fact. The last part, for me, has been key to building it in my career (30 years in infra/tech/security).

In favor of the generalist, I think AI is also quite limited in its scope of how it generalizes. I'm mowing through hundreds of mythos-generated security findings right now for work and while it's amazing that it can build an exploit chain 20 steps deep, it's completely lacking in all of the external layers that render it's speculation moot.

I've made an entire career out of being 'jack of all trades, master of none'. Being able to synthesize connections from relatively trivial knowledge in a bunch of domains is SOP for many humans as well. I think AI just has deeper knowledge and better pattern matching to make up for it's (at least now) lack of strength in cognition and 'ex nihilo' creativity.

(Which probably isn't 'ex nihilo' at all, and has more to do with the plethora of modalities that humans live in vs. large language models. For example, why do we pick the color red for notating important things and why do we say a schedule 'slips'...these are informed by a shared human experience borne of distinct physical sensation deep in our wiring that LLMs can only infer from what we write.)

I had no background in dsp or signal processing and found it so opaque as to be unusuable, even though I saw more skilled folks do amazing things with it.

*scratches head*

Might give it another go.