corvus-cornix

Some of the formatting for emphasis and em-dashes looked a bit iffy. It's hard to imagine AI wasn't involved, at least an an editor or to discuss ideas. To be clear, I don't think there's necessarily anything wrong with this.

I feel like this comment is channeling https://motherfuckingwebsite.com/

I feel like Go has a similar role to Java. Although it's mercifully free of inheritance and the functional stuff they've bolted on.

Rust has a similar role to C++ but reads more like Python and Elixir's lovechild.

I'm looking at engineering job specs at the moment and it's very wearisome that every company seems to have pivoted from highlighting the unique value they provide to customers to putting AI front and centre in their employer branding. My eyes immediately glaze over at what may have been the result of "Claude, take this HR/marketing/whatever copy and inject some AI".

I've adopted the tools because they're useful, but businesses need to chill. AI seems to amplify existing bottlenecks within organisations, so we should probably tread carefully when it comes to pushing the tech. Fix the organisational problems first and hedge our bets.

I wonder if anyone reading this was around during the dot-com bubble because maybe it felt the same...

Data recently scraped from Glassdoor and Crunchbase.

This essay examines the LightFM recommender system through Deleuze and Guattari’s concept of micropolitics. It argues that machine learning models like LightFM don’t just reflect user preferences—they help produce subjectivity by embedding users in global infrastructures of control. Drawing on theories of the machinic unconscious and proto-subjectivity, the piece explores how prediction and vectorization reshape agency, identity, and power in subtle, material ways.

I've also found LLMs to perform poorly at DevOps tasks. Perhaps there's a lack of training data. On the bright side this hints at better job security for platform engineers.

Goal: Setup a minimal working example where pods in GKE and EKS can access two buckets, one in S3 and another in GCS with no static credentials or runtime configuration. In other words, running the below commands should just work from within a pod in both clusters:

I only use notebooks when I need to analyze a csv and a python repl session would be cumbersome. For this limited use case I think they're fine.

Bash one-liner that produces a leaderboard of TODO frequency for an entire repository

Data including salary ranges, funding stats, employee ratings and tech stacks of tech (and tech-adjacent) companies that offer remote work.

They never mention the economist he collaborated with in inventing game theory: Oskar Morgenstern. There were also collaborators involved in developing "Von Neumann" architecture. He was just the biggest name involved. He was a genius, but there's some politics at work in who is remembered for collaborative intellectual work.

I'm sure I've seen the opposite of this recommended: https://www.django-antipatterns.com/antipattern/over-use-of-... Why pass objects with DB access into a template? I guess some people enjoy debugging templates and n+1 problems. (edited)

Learn FastAPI instead (for new projects where you don't want the template-based MVC). Django is showing its age and comes with some unpalatable design decisions, particularly in its active record ORM. Django's most useful feature is arguably automated DB migrations, but this can also be achieved with SQLAlchemy+Alembic