AI - AI Dynamics

AI Pen for Students: A Dream Tool or an Examiner’s Nightmare?

By

–

29 March 2026 3h17

LMStudio is my easy way of running llama.cpp kudos to @yagilb

→ View original post on X — @petergostev

Codex AI Model Code Analysis Reliability Discussion

By

@petergostev

–

29 March 2026 1h54

Ha for sure, I believe you – I had Codex look at some of these and they definitely looked quite sus

OpenAI Employee Claims 80% of Code Now Written by AI

By

–

29 March 2026 1h52

Nice list I think it’s about time-especially with all the expected openweights-to update my list of recommendations

→ View original post on X — @bcherny

Working at Anthropic: Daily Unusual Requests

By

@bcherny

–

29 March 2026 1h51

A weird part of working at Anthropic: getting a few of these each day

→ View original post on X — @hardmaru, 2026-03-29 00:46 UTC

Growing Up in the 90s: Neural Nets, Scaling Laws, and Artificial Consciousness

By

@hardmaru

–

29 March 2026 1h46

My ideal timeline: Growing up in the 90s, discovering neural nets, scaling laws, and building an artificial consciousness.pic.twitter.com/FQMunONGm3
— hardmaru (@hardmaru) 29 mars 2026

My ideal timeline: Growing up in the 90s, discovering neural nets, scaling laws, and building an artificial consciousness.

High expectations for Claude model despite recent setbacks

By

–

29 March 2026 1h35

We’re all doing our best Hopefully I’ll make local inference the easy by-default within the next 3 years (also why I am raising haha)

→ View original post on X — @linusekenstam

Vibecoded Browser App Simulates Electronic Circuits

By

@linusekenstam

–

29 March 2026 1h34

Someone vibecoded an app that allows for you to simulate and check your electronic circuits in your browser 👇 pic.twitter.com/UHFPMPNlbo
— Linus ✦ Ekenstam (@LinusEkenstam) 29 mars 2026

Someone vibecoded an app that allows for you to simulate and check your electronic circuits in your browser

Learn to Use AI: Training Courses Available

By

–

29 March 2026 1h33

Gonna give them a try tomorrow

→ View original post on X — @alexjc, 2026-03-29 00:25 UTC

Three-month-old Anthropic model achieves SOTA on code maintainability

By

@alexjc

–

29 March 2026 1h25

~3mo old model is still SOTA and was +8% vs 5.3-codex on maintainability total anthropic victory Gabe Orlanski (@GOrlanski) We found that agents generate progressively worse code with each iteration. Real developers do not. SlopCodeBench is the only eval that faithfully measures quality degradation on iterative, long-horizon coding tasks. arxiv.org/abs/2603.24755 scbench.ai 🧵 — https://nitter.net/GOrlanski/status/2037560777356238881#m