Search for a command to run...
Just yesterday, Bryan Cantrill published a short piece that touches something that I think most of us fundamentally understand, but don't have the words. The article is very short, titled "The peril...
When talking about Property-Based Testing, we typically talk in very abstract terms. There are properties, which define the correctness; there are generators, which define the domain; the PBT...
This is a mirror entry of an article I co-authored published in Datadog AI Blog. AI agents can now produce software faster than any team can verify it. The bottleneck has moved from writing code to...
This is a mirror entry for my article published in The Consensus. Welcome to a snapshot of my multi-year long journey of inferring types from jq programs in order to produce better, understandable...
Almost a year ago, I wrote my first ever blog post on AI-assisted programming, Verifiability is the Limit. The core idea was a push against the school of thought that AI-assisted programming would...
2026 started with a boom of AI-assisted autonomous translations, on 14th of January, Cursor published their post on Scaling long-running autonomous coding in which they created translations of a...
I’ve been going round and round in my mind about a particular discussion around LLMs: are they really similar to compilers? Are we headed toward a world where people don’t look at the underlying...
AI is making formal verification go mainstream. AI-assisted mechnical proving companies are raising funds on billion dollar valuations, new people are trying proof assistants, overwhelmingly Lean,...
Hatalar, günümüzde programlamanın ve yazılım mühendisliğinin ayrılmaz bir parçası. Yazdığımız programlarda hatalar olacağını, bu hataların kaçınılmaz olduğunu, bu...
Random testing, in isolation, is a stupid idea. You take a program, which has a practically infinite amount of possible inputs, you randomly pick one of those infinite possibilities in the hopes...
This post is a sequel to my thoughts on "What are the limits of LLMs" (Verifiability is the Limit) and "What are the domains I believe LLMs will shine at autonomously" (Breaking Verifiable...