mfrye0

Does anyone know how this compares to Carta?

While not a site with user generated content, our version of this was a huge increase in spam accounts.

We offer an API with a free tier. So naturally people would create multiple accounts to avoid having to pay.

It's been a huge lesson in rate limiting, IP blocking, and verifying accounts.

It was a 3rd party resource that when updated would call a lambda via a webhook, which would then update the 3rd party resource. So it would create an infinite loop for each resource that was modified.

Dumb mistake...

I was able to get AWS to forgive the majority of the bill. I think that was my one time pass though.

I believe so, if it's a T series. For 100% CPU usage, it's cheaper to just get a non T series instance like a M.

One of the biggest mistakes I made is not exploring spot instances and reserved instances earlier.

I cut my bill by 70-80%% after paying full price for years...

If you have an active web server or backend workers with fairly short jobs, spot instances will work for you.

> If you want to talk systemic AWS mistakes you can make, we accidentally created an infinite event loop between two Lambdas. Racked up a several-hundred-thousand dollar bill in a couple of hours.

I did more or less the same thing, but with a 3rd party webhook. The bill almost killed my company.

Got it. Yeah, it's definitely tricky.

The other aspect is that a decent chunk of the IPv4 space at least is fairly dynamic. We've seen some blocks change owners every few weeks.

I've been working on this and have built that database, though we only expose at the IP level: https://bigpicture.io/docs/api/#ip-api.

What did you have in mind as far as a use case?

I haven't heard of this thing yet. Do you have one?

Yeah, I was in a similar place at one point. Check out the book in this thread: https://news.ycombinator.com/item?id=6099623. The book is: https://www.amazon.com/Mindbody-Prescription-Healing-Body-Pa...

I don't know if it was the book, a placebo effect, or something else, but I got better after reading that.

I was on the list for a SDK at CTRL-Labs before FB swooped in and bought them. I'm really bummed I couldn't get ahold of one first.

In having on-again, off-again RSI issues over the years from coding, I think this tech would be hugely beneficial and is more realistic to ship in the near future.

I've been doing some AWS optimization recently. For my setup, which heavily uses a lot of short job based queue patterns, I've had a lot of success with the following:

1) Moving to spot instances where possible 2) Autoscaling rules 3) Reserved instances where possible 4) Moving everything to AWS Graviton based instances

I cut the bill by ~70%.

This is great. I've been thinking about hacking something like this together for myself for awhile now.

Part of my interest in this is that I'm sitting on a bunch of company data already - both public and private companies. I'm the founder BigPicture, a B2B sales tech startup, where we had to acquire similar data sets to power our product.

Basically, we got sick of paying Clearbit for the data, so we ended up building our own thing. Here's our API docs that shows a sample of what we have: https://bigpicture.io/docs/enrichment/company/

Getting this data is a pain in the ass, so if you're interested, I may be open to partnering in some capacity. My email is michael [at] bigpicture.io

> hearing that this is your third project in the past 8 months is not exactly what I want to hear from my B2B provider.

Ironically one of the top live chat companies right now was project 7 or 8 within a year before settling on chat.

One thing that threw me off is upserts with NULL in Postgres. NULL != NULL so it will continually create new rows.

I went through a similar situation and ended up using https://voicecode.io for a few years. It worked well enough for me, but I always found it hard using it in a public setting.

Ironically, I solved my RSI issue after coming across a book on HN. Everyone seemed to recommend the Mind Body Prescription, and after trying it, it did the trick.

https://www.amazon.com/Mindbody-Prescription-Healing-Body-Pa...

Edit: I've been wanting to try this project for several years now: https://www.ctrl-labs.com. I think something like this is the solution moving forward vs voice coding.

I had a friend tell me recently about a client using commercial real estate data for lead gen. He mentioned https://compstak.com/

Basically, identifying companies that are doing well / expanding by how big the space is they leased. This sort of data is apparently very hard to get, but gives users a competitive advantage.

Yeah, he didn't do the best job of explaining why in that example.

Using a global store makes sense when your todo list evolves and the component storing the data is no longer mounted / exists. Examples could be:

- Full page individual todo edit - Separate component / page that uses the same data - Some background logic that does X based on the todo data

For anyone not familiar with UI app design, you do not need to always default to a Redux or Mobx. I've found a simple object / class with an event emitter will suffice for many cases.

Not necessarily.

If you fail, just get a job, save up more money, then try again.

The more you fail, the more you learn, the greater the likelihood of success the next time around.

Are you? Never heard of that, but it looks pretty cool.

I've been listening to binaural beats lately at the recommendation of a few friends. It seems to work pretty well to drown out the office noise without distracting me.

Can you elaborate on "BERT + rules + a classic search engine"?

We're due to upgrade our basic Postgres based text search to a formal search solution and investigating different options.

You beat me to it. I was going to share this too.

I believe Michael Seibel had some choice words on Twitter last week about this guy.

If you want to see one of the interviews, the one from Workato is particularly memorable: https://www.youtube.com/watch?v=Cg8q6RqpefM. The founder couldn't handle the pressure.

Agreed. We've been using this successfully for awhile now.

I'm curious though at what point something like this or ES itself would make sense for primarily text search. Is speed the biggest thing, or is it more flexibility to tweak and get better search results?

I've been using Dbmate for a few years now across our entire stack. It does the job perfectly.

https://github.com/amacneil/dbmate

Congrats on the public launch. Some of our customers have been trying you out over the last year and I've been hearing good things.

When things calm down from the launch, we should speak further. We built / released our own Company and IP APIs since Clearbit was obscenely expensive at scale. Our data quality / match rates are now comparable - https://bigpicture.io/docs/enrichment/company/

We're looking for a couple strategic partners. You can reach me at michael [at] bigpicture.io

That's a good idea. I think a blog post explaining the strategy further could be helpful for a lot of devs.

To be frank though, I really don't have time at the moment. My typical blog post takes a few days to write, and I'm completely swamped with work for the next couple months. Like 12-15 hours a day, no life sort of thing...

With that said, I'd be open to chat with you or someone else about it, and they could run with it on their own.

My email is hello --at-- michaelfrye.me

Because most of the code you write at a previous job is not going to publicly available. So I have no other reference point as to your current skills.

Apart from just the code itself, having side projects is a good sign of valuable, non-technical skills. Such as intellectual curiosity, ability to start/finish things, and if it's a collaborative open source project, can play well / collaborate with others.