jarl-ragnar

I think this misses one of the key points of Elixir, Phoenix and LiveView. How do your benchmarks look when under load? How does each handle 1000 concurrent users or 10,000 etc. In Elixir each LiveView has its own process, how does Django handle that?

We binned any form of coding questions or take home task. Our preference now is to focus on architectural concepts and ability to apply those to a problem.

So, as happened last week, if I’m interviewing for an Elixir dev I’m going to be interested in your knowledge of the BEAM and how it’s features can be used to solve common architectural problems.

Maritime vector charts for use in mapping applications https://marinecharts.io

Current coverage is the US, more countries coming soon.

It is. I’ve found it so useful I created an Elixir library to act as a cache in front on a pmtiles file. https://hex.pm/packages/ex_pmtiles

I just use Quickbooks. Takes a little getting used to but I can just give my accountant access and they can get what they need to prepare the end of year accounts for filing.

There was a similar project in the UK back in the 90’s aimed at using remote sensing to detect mass graves in Bosnia.

A mix of cursor and claude. I’ve just used cursor to help me write a radar data visualisation tool using SwiftUI and Metal. Never used either before.

In essence I treated it like a junior dev who already knew the tech stack. So tell it what I wanted to achieve then pair program to get it working. Then iterate to add functionality. Once a feature is working, I could then ask it to help me understand how the tech stack was being used. Lots of fun, and very productive.

I followed a similar approach recently to create map layers with Sentinel-2 imagery.

Webp encoded tiles extracted from the CoGs available on AWS then embedded within a pmtiles file.

I was seeing almost an order of magnitude reduction in file size compared to the CoGs with no visible image degradation.

I’d recommend listening to Lenny’s podcast.

Clojurescript. I used it to build an MVP proof of concept for work and now have to watch a small team re-write it using Typescript and Angular

They’re still not at feature parity with 2x the team, 2x the time and 3x the lines of code.

Sounds as much a problem of too many Product Managers all trying to make a mark so they can progress their career. Hence features keep getting added or re-designed.

Completely agree. Though I’ve also found an Elixir backend works well because of the power of the Beam VM.

That’s pretty much the destination I’ve arrived at though in my case via the Elixir route.

I’ve been impressed with Elixir. At my current gig it’s easily replaced a bunch of java microservices using RabbitMQ as the messaging backbone that struggled to handle the workload. To try and handle the load we kept upping the hardware spec.

I thought that Elixir could do better and created a replacement Elixir monolith. The replacement happily handles the workload while running on my MacBook Pro.

I did something similar to this last year. Went from site Engineering manager at a multi-national aerospace company to IC role and don’t regret it.

I did resign, and started in a software architect role at a smaller company. Did that for a bit and then moved to become their lead data scientist.

I was in the eng. management role for nearly 10 years but kept my skills relevant by working on lots of side projects.

Trying to manage a team remotely through lockdown and the never ending dance of aerospace mergers and acquisitions followed by the inevitable reorganisation was largely what prompted the change.

Can’t argue with that assessment.

Elixir and Phoenix Liveview. It really is the most productive framework I’ve found.

Watching slightly incredulous as a newly elevated CTO instigates Scrum, I was drawn back to this gem.

Any large aerospace company. Especially those that have been bought and merged with others as the sector has consolidated over the years.

Having watched it from the inside of one through 4 aquire/merge cycles over 10 years I speak from experience.

All project management is risk management

https://www.infoq.com/presentations/risk-project-management/

Know what you’re building but risk of schedule and budget overruns - waterfall and possibly earned value management (loved by the defence and aerospace sectors).

Not sure what you’re building so risk is you build the wrong thing - agile, get something minimal into the user/customer hands and course correct based on feedback.

No individual framework is perfect.

Start by reviewing your company process for performance management. Then follow it. At each step notify HR of your intentions. Document all meetings you have with the individual concerned and HR.

Set clear achievable short term goals with the individual that enable you to clearly identify and articulate their poor performance. Use that to reiterate performance expectations. If performance doesn’t improve, begin escalating with HR support (by that I mean tell them what you intend to do next and get them to sit in on formal meetings).

Having been through this myself it’s about ensuring it isn’t a surprise to the individual concerned that their performance is poor and creating the evidence that you followed a reasonable process leading up to any termination because worst case you and the Company end up being taken to employment tribunal. And that’s where your thorough notes and evidence that HR were advised on the steps you were planning to take are key.

And as a manager you have to pursue this because the wider team will know who’s not performing and be watching to see how the business addresses it. If you’re lucky the individual will decide to leave once they realise the direction of travel.

Good luck. And I recommend the book “Radical Candor”

I’d encourage anyone who sets high standards for themselves and others to listen to this from Jonny Wilkinson https://podcasts.apple.com/gb/podcast/the-high-performance-p...

This rang so true. I recently left a large multi-national aerospace company where I and my team had developed local processes that put systems engineering first.

Unfortunately our main contract was developing a component of a larger system being developed by our parent organisation who didn’t have a concept of systems engineering. We tried for years to educate them and I watched aghast as their program costs and schedule continued to spiral out of control.

Basically a bomb burst of engineers all doing what they thought was the right thing but no one owning the system design and saying no to good ideas.

In the words of their chief engineer “its like 10 different black boxes, I don’t know what I’m getting and I don’t know when it’ll be finished!”

3 months in as the “old guy at a startup” and I can concur with a lot of this. Especially that my emoji game is weak.

I’d highly recommend Most Secret War which covers the role of Scientific Intelligence in Britain’s battle with Germany during WWII.

https://www.amazon.com/Most-Secret-Penguin-World-Collection/...

No idea. I think the demo schedule drove him to find something that he could get working.

To be fair, I think he was fighting not just Kotlin JS but using React within it and his own lack of web technology experience being predominantly an Android dev.

One of our lead developers at work pushed this for a rush demo job last week (he’s a Kotlin advocate). He spent the week fighting with lack of documentation and has now given up and reverted to plain old javascript.

I’d just use Clojurescript.

Great article, the intro captures the core reasons why I recently quit as an engineering manager at a multi-national aerospace company. It also rather succinctly underlines how Boeing end up with the 737 Max situation.

In my experience, internally funded research and development. A proportion of revenue is reinvested to fund IR&D projects focused on areas identified through longer term engineering roadmapping activities.

An end to companies laying claim to IP generated in your own time.