
Joe Magerramov
@_joemag_
Followers
4K
Following
3K
Media
106
Statuses
3K
VP/Distinguished Engineer at AWS. Misplaced New Yorker in PNW. Die-hard sports fan on hiatus.
Joined February 2019
I find that there are simple and self-contained coding tasks that can be completed with just a series of prompts. For those, the simplicity of Claude Code or Q CLI is actually an advantage over the feature richness of an IDE.
I don't really get Claude Code. Why do you guys want to sit in a terminal to look over changes vs the multi-tasking capability and UX of the whole IDE? I must be missing something!.
0
0
5
There's the other extreme too! While reviewing the metrics for Amazon Bracket (AWS service that lets you run tasks on quantum computers) we joked that given their uniqueness and scarcity, they are neither pets nor cattle, but more like family!.
There needs to be some philosophy that's in between pets and cattle. I nominate Furbys. There's still loveable and cute and you care about them, but there's a lot of them and they are pretty easily replaceable.
0
0
9
I'm biased, but @LIGO has to be the most impressive feat of human engineering to date! . Detecting Black Hole mergers from billions of light years away by measuring spacetime changes of 0.0001% the width of the proton! 10^(-20) meters, hard to even wrap your head around!.
Nearly 10 years ago, the #NSFfunded @LIGO first detected gravitational waves — actual ripples in space-time! Today, it continues groundbreaking work with upgrades to detect more subtle waves & expand our understanding of the universe. 🌌. 📷: LIGO
3
3
12
RT @spikedoanz: they don't want you to know this, but you can visualize arbitrarily hyper dimensional spaces in your head. just keep stack….
0
998
0
Lot’s of options here, but my favorite way to win this game is not to play it! As an alternative to read-through caches, I’m a big fan of materialized view caches. They work by materializing the cache from the change log produced by the source of truth. 1/.
In a distributed environment, one of your cache layers (e.g., Redis) fails. As a result, all requests hit the database, causing a cache stampede, where multiple microservices simultaneously query the database for the same data, leading to a bottleneck. How do you design a.
1
0
12