How the Lobsters front page works
(atharvaraykar.com)
Stories by atharva
JetBrains Mono: A free and open source typeface for developers
(jetbrains.com)
Absurd Workflows: Durable Execution With Just Postgres
(lucumr.pocoo.org)
Write code top-down
(teamten.com)
The Resonant Computing Manifesto
(resonantcomputing.org)
Don't Post Passive-Aggressive Webpages
(dontpostpassiveaggressivewebpages.com)
How I keep up with AI progress
(blog.nilenso.com)
Readable Code is Unreadable
(blog.wilsonb.com)
What are popular AI coding benchmarks actually measuring?
(blog.nilenso.com)
An RSVP speed reader for your terminal
(github.com)
Expensively Quadratic: the LLM Agent Cost Curve
(blog.exe.dev)
My Quarterly System Health Check-in
(blog.nilenso.com)
The Problem With LLMs
(deobald.ca)
Artisanal shims for the bitter lesson age
(blog.nilenso.com)
How StrongDM’s AI team build serious software without even looking at the code
(simonwillison.net)
Paper Abstract: Large language models (LLMs) are increasingly used as evaluators in lieu of humans. While scalable, their judgments are noisy due to imperfect specificity and sensitivity of LLMs, leading to biased accuracy estimates. Although bias-correction methods exist, they are underutilized in ...
Why BAML?
(docs.boundaryml.com)
Small Models Can Introspect, Too
(vgel.me)
Open Responses Specification
(openresponses.org)
Minimum Viable Benchmark (for evaluating LLMs)
(blog.nilenso.com)
Building Repo Bench
(repoprompt.com)
How We Made SWE-Bench 50x Smaller
(logicstar.ai)
Why Does AI Feel So Different?
(blog.nilenso.com)
I Taught My Dog to Vibe Code Games
(calebleak.com)
Weird system prompt artefacts
(blog.nilenso.com)
Optimizing Volumetric Particles with AI-Assisted Coding
(jysandy.github.io)
Nemotron 3 Nano Technical Report
(research.nvidia.com)
This particular model release is interesting because not only is it open weights[0], but they also released most of the pretraining and posttraining data along with the code and libraries for the RL environments (much like Olmo3, which came out recently). It is also trained from scratch and architec...
How to work with Product: Taste and Adjust
(blog.nilenso.com)
Fight context rot with context observability
(blog.nilenso.com)
Data on AI Models
(epoch.ai)