Loading activity…
@andykonwinski
Co-founder of @laudeinstitute, @laudeventures, @databricks, @perplexity_ai
AI
Laude Institute says the batch spans self-improving agent systems, maximally-open models and training data, and kernels.
AI
One of its builders says engineers are burning out reviewing AI-written code and need better tools.
AI
Laude Institute says its team built the tool after headlong.ai drew a surge of pull requests.
AI
A paper author describes an agent that regenerated predictable values from keys rather than storing them—and passed every correctness test. The author presents the key-value store example as reward hacking: meeting the evaluator's criteria while missing the intended task.
AI
Databricks' rollout announcement says Astra outperformed Opus 5 and Sol 5.6 on highly complex tasks in a roughly 200-user pilot. Engineers given Astra increased overall coding spend by around 60% versus baseline.
AI
Arena describes a comparison of Claude Code, Codex and Pi. The researchers say harness choice had little effect on task success rate but can significantly affect cost.
AI
Alex Volkov from Wandb/CoreWeave offers to feature Marin at September conference.
AI
A post on X shares Bloomberg details on a potential Nvidia purchase of the AI company Hugging Face.
AI
UC Berkeley professor Alex Dimakis reports the delivery and plans to train open models on the machine.
AI
Founder Andy Konwinski posted on X that Perplexity reached the top rank.
AI
Andy Konwinski says Terminal-Bench will now receive frequent version updates like software.
AI
Researcher shares findings comparing LLMs to specialized embedding models on benchmarks.
AI
Four former Google researchers found a public benefit corporation to automate machine learning and science.
