epberry

I don't necessarily think that billing which is hard to estimate is good for anyone. For large deals I've seen this be a real turn off. Same with surprise bills. Ultimately if the service can scale billing in a way that their value scales to the customer (Datadog and Snowflake come close to this) then both the provider and the customer will be happy.

I think the market for automation and workflow building for non-coders has just grown tremendously. Zapier at this point is probably also focused on winning enterprise customers leaving plenty of room for others to steal users for specific use cases on the low end. Finally I believe Zapier raised their prices at the end of last year (or at least many people got an email to that effect).

Startup = Growth is timeless, as are most of the essays. I did feel Hackers vs Painters was little outdated but the book is almost 30 years old at this point.

We just switched to DuckDB from Postgres to analyze AWS billing data and wrote up our experience, https://www.vantage.sh/blog/querying-aws-cost-data-duckdb

Arguably Postgres was never the right tool to use for this analysis but nonetheless I was surprised at how much faster DuckDB was.

That is crazy to think about. I wonder what this sentence will be about in another 25 years.

I would like to contribute to rewilding projects like https://www.americanprairie.org/ and build software to support them or if possible buy land directly for them and establish more throughout the US.

Absolutely agree with the love for the OG homepods. Great sound and I use it many times daily. Finally got my Hue lights hooked up with Homekit too so that is working well. Lastly it is my TV speaker for my Apple TV.

I've tried this out and it's very cool. I think one particular application is roll-your-own "social listening" which usually requires buying tools.

My favorite take on pricing ever is "Index to your Customers' Revenue Growth" (fourth card on this page) https://www.screenshotessays.com/browse. Basically saying that if your cost grows with your customers' revenue or some first derivative of that then you will be set for life.

> Currently much of our database access is ad hoc with a proliferation of Perl scripts that to a very real extent run our business.

There are companies started later than 2010 where this was still the case. Interesting to think about how shipping things quickly is so different than scaling them up.

The main factor is level of spend. If you can commit to spending $1M over the next 2 years AWS will give you big breaks.

Very nice write up. As a billing guy I especially liked the tip about using resource policies to enable cross-account access and save on KMS request costs.

There's one issue with companies using hundreds of AWS accounts if you're a vendor to them: integrating services. Some folks here may be interested in a technique called "CloudFormation StackSets" which can deploy bits of infrastructure to multiple AWS accounts in one command. Vantage uses this to setup our billing integration and we wrote up the method here, https://www.vantage.sh/blog/using-cloudformation-stacksets-t...

Thank you for building this! Networking is one of _the_ hardest things to price out in cloud billing.

As a maintainer for https://ec2instances.info I absolutely love sites like this. Nice and clean. Just like that site this is statically rendered into a large HTML file and delivered to the client, with data updated every 2 hours via scraping.

And sure enough if you search 'disk prices' or 'harddrive prices' this is the #1 result which I am certain helps with their Amazon Affiliate revenue.

I like the 'internal foundry' messaging. It feels cool and futuristic. Also given recent tensions in the South China Sea it probably makes sense to get that going in a serious way for national security reasons.

I work for Vantage and I do think that startups are starting to care about cloud costs, but not necessarily for the cost itself.

The thing to understand early these days is COGS and whether the business can scale its service with good margins. That is a significant change of concerns for investors from the past years where growth was the only thing that mattered.

The other thing that I think is a little off in this post is the idea that startups not caring about costs are a big driver of hyperscaler revenue. I suppose that may have been true a few years ago but AWS, Azure, and to some extent GCP are having a lot of success moving legacy on-prem deployments to the cloud. It's not uncommon to encounter 50 year old enterprises with a billion dollars of cloud spend. One example of this is the dedicated SAP instances that AWS has, some of which cost $100K per month.

I can tell you that NASA is in the midst of a multi-year effort to move their computing to AWS and that yes.. downloading 324 terabytes of data is very expensive but very soon all this data will just remain in the cloud and be accessed virtually.

Great stuff! I know the ClickHouse team and they are world class. If you're more familiar with a relational database some things will feel weird. For example, you should not insert 1 row at a time, there's over 1,000 built-in functions, and the default connection is often over HTTPS, but not 443.

Multiple AWS accounts is definitely a best practice in larger engineering orgs. We implemented a CloudFormation StackSet to ingest them into our billing tool and lay them out appropriately. This proved to be a very slick solution from AWS so it made me believe that they want their larger customers to use multiple accounts too.

If you are smaller I would not recommend it. Many things become a little more difficult, as others have pointed out. Oftentimes a devops or platform engineering org will paper over these things.

Yep they are extremely aggressive. The technology is also extremely hard though.

I believe this is the use case Cloudflare is really targeting with R2. They recently connected Cache Reserve to R2 to make this even easier. We wrote up a breakdown for S3 vs R2 and found that R2 would be significantly cheaper when the majority of traffic is cached data, https://www.vantage.sh/blog/cloudflare-r2-aws-s3-comparison

Yeah absolutely. Another huge thing I got some window into at Qualcomm is technical validation. So Apple is going to be looking for proof that the chips deliver and TSMC is going to be pushing back against high expectations on that front.

Incorporating tags is really nice. I appreciate infra companies that help teams stay organized with their resources.

Ditto is showing what is possible with the edge. Very excited for this and all the new emerging architectures.

Echoing this. The SRE book is also highly revealing about how Google request prioritization works. https://sre.google/sre-book/load-balancing-datacenter/

My personal opinion is that Google's resources are more tightly optimized than AWS and they may try to find the 99% best allocation versus the 95% best allocation on AWS.. and this leads to more rejected requests. Open to being wrong on this.

This finding could shed more light on the timing of the so-called 'bronze age collapse' which Jared Diamond and Eric Cline (author of 1177 BC) have popularized. Cline's conclusion is that the collapse began around 1230 and continued on for 50-100 years throughout the mediterranean.

Multiview, multi-object tracking - https://github.com/prcvlabs/multiview

I started doing research on cameras around 2012 and have been obsessed every since. The company behind this code failed unfortunately but I've kept poking at it and finding more people interested in it, even as my day job is more about cloud infrastructure. Companies like Zippin which have carved out a niche in the cashierless space really impress me.

I have found this to be even more true in non-technical fields. 2 blog posts a week vs 1 blog post a week is the difference of tens of thousands of hits on Google in a year.

You could join a startup. It won't solve the always-on problem but it will certainly solve the bits with legacy code and not creating new things. You can also look for a place that is in-person which could give you some highs of working with a designer, working with senior engineers, and in general feeling part of a problem solving team.

Some of these might be good - https://www.workatastartup.com/. I also think Turing is placing engineers well. Not sure about TripleByte.

I participated in another Daniel Gross program, Pioneer, and it was _excellent_. I highly recommend it or this for folks that may be more 'outside the band' when it comes to raising money or getting a startup from 0 to 1. I would note that if you're trying to bootstrap this is probably not the program for you. These guys are incredibly connected in venture capital and this program ought to be a great way to leverage that.