twistedpair

There is a company that delivers broadcast video ads to hundreds of TV stations on demand. The ad has to run and run now, so they cannot tolerate failure.

They write the videos to GCS storage in Google Cloud, and to S3 in AWS. Every point of their workflows are checkpointed and cross referenced across GCP and AWS. If either side drops the ball, the other picks it up.

So yes, you can design a super fault tolerant system. This company did it because failing to deliver a few ads would mean lose of major contracts.

Did this affect a lot of tech folks, or other non-tech front office/sales positions? Unfortunate close to the holiday.

mabl | Boston, MA | Software Engineer | ONSITE/REMOTE | https://mabl.com/careers

We need your help to put our Nov $40MM series C raise [0] to work and keep scaling up. We’re bring AI and cloud scale to software testing at mabl. We provide automatic, self healing testing of web applications using advanced ML and AI techniques to replace the tedious and fragile UI testing of yore.

We’re a 100% serverless operation built on Google Cloud Platform that rapidly develops and deploys features on a CI/CD model. Four years in, we’re a recognized industry player and growing our engineering ranks to keep building out our platform.

Our open positions:

- Software Engineer

- Software Quality Engineer

- Product Manager

- Inside Sales Engineer

Our stack is built with modern Java 11, TypeScript, and some Bash glue

Drop me (an engineer), any questions joe at-symbol mabl.com, and checkout our careers site [1]. We can’t wait to work with you.

[0] https://venturebeat.com/2021/11/18/mabl-aims-to-automate-sof...

[1] https://www.mabl.com/join-the-team#jobs

I learned web programming on PHP, and a ton of anti-patterns well. Making a single script with a very rich stdlib is quite handy, as well as "just running it" on your desktop instantly. However scaling out large projects in a performant way, doing big refactors, or mature server programming, I'd pick nearly any other tool today.

Disclaimer: I still maintain my PHP site from '08, so I'm current on trends.

With great power comes great responsibility. I've seen Spring projects where DI was a total rat nest of hundreds of XML wireup config files. Downhill with a tailwind, it just might boot correctly, after minutes classpath scanning for dependencies.

Use DI judiciously where it provides the most value. Type, if you can type it (e.g. I vastly prefer explicit wireup in Guice Module.java files over Spring's "stringly typed" XML files. Type checking is more likely to catch classpath errors and typos. Event better, use compile time macro wireup like Scala's Macwire.

Common use cases: if DI can inject different implementations, you can have a PostgresDaoImpl and SqliteDaoImpl. The DI config for running in prod uses all the Postgres implementations, while the impl for running unit tests locally uses the Sqlite impl.

Your code stays clean, because nothing on the consumer side needs to change if in prod/test or postgres/sqlite. Better still, your code doesn't even need awareness that prod/test environments even exist. Keeps things simpler in the long run.

Watched my neighbor powersand lead paint off his house yesterday. He wasn't even wearing a mask. Makes you wonder how many decades until we don't have as much lead in the environment? 50 years?

+1.

I've done cross cloud connects where the cloud vendors are on the phone with me and eventually say "We really don't have any idea why the tunnels are not connecting". And that's when using Cisco VPN appliances, no less, following everyone's setup instructions to the letter.

Indeed, not a trivially solved problem yet.

Maybe big oil and old-auto are the bigger competiton for Tesla, so it's best to have other automotive heavyweights going all in on EV.

How often does the CEO of one company address management of its biggest competitor? How often can you even get said CEOs to public acknowledge the name of their competitors?

Elon started Tesla to bring about the EV revolution. He's not throwing dust in the face of the competition, but rather telling them how to catch up, because he wants their joint goal to be realized: ubiquitous EVs.

I think the bigger picture is _sleep hygiene_. Don't use your LCD device before bed. Don't have all your birght LEDs blaring in your house until midnight.

We evolved as diurnal monkeys, with bright days and moonlit nights, staring at a mix of near and distant focal points all day. Now we stare at a glowing blue rectangle 1-2' from our faces for the majority of waking hours.

Break the cycle. I for one read (paper books, or on tape) after 9pm. I have zero devices, clocks, LED's, etc in my bedroom. My Phillip's Hue lights step down and dim as the evening progresses, rather than lighting my house like a bus station. I sleep much better than I used to when I sat in front of 3x blaring LED monitors until 1pm each night.

Good luck.

More details: https://stockoptioncounsel.com/blog/standards-ownership-cant...

To clarify: first engineer.

Also watch out for a clause giving the buyer the right to buy back the employee shares, possibly at par value.

That means you could have executed those options 5 years ago (to prevent a windfall tax), and just get you money back at acquisition time. Zero interest five year loan. Thanks!

Good point. I left my last startup for this reason. Kept telling the CEO "show me the money". After a half dozen times being told that the "funding is just around the corner" I left.

Two months later they cut more than half the staff and becoming an IP bundle looking for a buyer.

Friend was the founding engineer at startup that did quite well. Eventually sold for 100s of millions, and yeah, he didn't see a dime.

Always discount those options to 0, and make sure you're getting a good salary.

Ever call Fidelity phone support and hear "enter your password on the keypad"? That means collapsing ~62 chars into 10 char options, a massive space reduction.

Then there's the fact that many banking sites (BofA, IIRC) only used the first 8 char of your password anyway.

Pretty sure Plato wrote about this in his _Allegory of the cave_ (c 400BC),

How many acres of trees does it take to become a "forest" and not a "grove"?

Phew, managed to quit FB in 2009.

So much for information compartmentalization. Does the typical engineer need access to payment details for their daily work?

I think this gets to the root of why you're leaving a tab open. You might have non-persisted state in the page, hence you don't want to close it.

MAD worked out pretty well. Sometimes the best outcome of a weapon system is that it's never used.

Reminds me of the Doomsday Machine in Dr. Strangelove.

> The Whole Point of the Doomsday Machine is Lost if You Keep it a Secret!

Ballistic missiles give of bright, very hot exhaust plumes on leaving their silos. Any major world power with a down looking infrared satellite (or watching for "WTF was that launch in my wheat field" tweets from farmers"), will know when it launches, and have a relative idea of where it's headed and when it will land.

A hyper-sonic cruise missile is much more stealthy, giving it the element of surprise.

The Sprint[1] missile accelerated at 100g. That's 0-mach 10 in 5s flat. It flew so fast in the lower atmosphere that it glowed white hot.

So, just saying there were some pretty fast missiles decades ago. There was also the HIBEX missile, with 400g acceleration.

[1] https://en.wikipedia.org/wiki/Sprint_(missile)

I'm not sure what this provides over what GCP already offers. 4 years ago I switched my co's ML training to use GKE (Google Kubernetes Engine) using a cluster of preemptible nodes.

All you need to do is schedule your jobs (just call the Kubernetes API and schedule your container to run). In the rare case the node gets preempted, your job will be restarted and restored by Kubernetes. Let your node pool scale to near zero when not in use and get billed by the _second_ of compute used.

Don't forget the Honda Jet, built in Greensboro, NC.

They have some aviation experience.

mabl | Boston, MA | Software Engineer | ONSITE/REMOTE | https://mabl.com/careers

Bring AI to software testing at mabl. We provide automatic, self healing testing of web applications using advanced ML and AI techniques to replace the tedious and fragile UI testing of yore. Consider us a smarter, auto-healing Selenium.

We’re a 100% serverless operation built on Google Cloud Platform that rapidly develops and deploys features on a CI/CD model. Four years in, we’re a recognized industry player and growing our engineering ranks to keep building out our platform.

Our open positions:

- Software Engineer

- Head of Customer Success, Head of PM

- Enterprise Sales, Account Executive, Sales Engineer

Our stack is built with modern Java 11, TypeScript, and some Bash glue

Drop me (an engineer), any questions joe at-symbol mabl.com, and checkout our careers site [1]. We can’t wait to work with you.

[1] https://www.mabl.com/join-the-team#jobs

I suspect they spend much more on transit than compute.

This is old school tech.

For example, I was tought how this worked in satillite image analysis optical computers (e.g. Cold War tech) when getting my optics degree.