Personal: Disciple of Christ & The Unix Haters Handbook Professional: Software Engineer. Burning tokens in "The Enterprise"
Joined February 2026
- Tweets457
- Following105
- Followers41
- Likes217
Batch Job retweeted
The 40-year-old COBOL code running critical applications isn’t tech debt anymore. Let's just pretend it was vibe-coded yesterday, and it becomes modern software.
Batch Job retweeted
Step back and think about the things women fight for harder than anything:
1. The right to murder their children through abortion
2. The right to have absolutely no accountability for anything, ever (see Lindsay Clancy
3. The right to be a whore (see Jane Doe/Cornell)
4. The right to dress like a prostitute everywhere (see Shane Gillis)
And their reactions to be called out for it exposes them further and further.
Look, ladies, the things you fight for most say a lot about you. It speaks to your values. And the point is... you don't have any.
*this is not every single women everywhere who has ever lived all the time, just most of you, so don't bother with the low IQ responses.
This paper from 1999 is interesting. Particularly section 3.3.3
chai.uni-hamburg.de/~moeller…
Batch Job retweeted
the bottom line is that there simply aren’t enough resources for ubiquitous AI driven software development. unless you are getting unlimited tokens from your employer’s own datacentres you have a responsibility to continue to own the software and be prepared for token scarcity.
also, this caveat every single “don’t read the code” advocate adds (“except for mission critical code”) is bullshit.
everything commercial is critical within its own sphere. be it the trust and reliability expected from customers, or the owners’ expectation that software professionals execute responsible stewardship while also delivering of business need.
this is true for the one-person business that pays the mortgage and feeds the family right up to the multinational that runs numbers for insurance underwriters.
Batch Job retweeted
Why? Why spend hundreds of thousands of dollars on slop tokens when you could direct the same effort and tokens at good, established, distributed projects like LibreOffice, Gimp, Kdenlive, Thunderbird, etc etc
My hunch is that this is pure clout chasing and they have zero interest in actually maintaining the software past its slop-formation
Service consumes the order request from a kaka topic patitioned on order ID. Then wrap the order processing in a DB transaction. Ack kafka when order transaction is complete.
Recently I interviewed a Senior Engineer.
Strong candidate.
Kafka? ✅
Redis? ✅
Microservices? ✅
Distributed systems? ✅
Then I asked:
"Your payment service processes a $49 charge for an order."
A customer clicks Pay.
The request reaches your payment service.
The card is charged successfully.
But before your service can mark the order as paid and send the response, the network connection drops.
The customer sees:
"Payment failed."
They click Pay again.
Meanwhile, the first charge already went through.
Now the customer has been charged twice for the same order.
How do you guarantee the customer is charged exactly once — even with retries, timeouts, and network failures?
"Just use a database transaction."
Doesn’t fully solve the distributed failure + retry case across services.
"Store the payment status in Redis."
What happens if Redis is unavailable when the second request arrives?
"Check if the order is already paid before charging."
What happens when two requests check the status at exactly the same time?
Your turn:
How would you design this payment API so concurrent requests and retries never double-charge the customer?
There’s a specific distributed-systems concept behind this.
And you’ll encounter it everywhere payments, orders, and financial systems are involved.👇
Batch Job retweeted
Your a Harvard PhD and NY Fed economist? wut
Unironically claiming Starbucks causally creates patents via latte chats, citing a non-peer-reviewed preprint with first-stage F-stats of 1–3.5 on 200 instruments. Starbucks opens where the money and talent already are. The instrument is weak, the effect size is absurd, and the mechanism is vibes.
If this is the standard of causal inference at the Fed, we have a problem.