Posted on

Jake Wharton announced yesterday that

Jake Wharton announced yesterday that all the Lysine projects (OkHttp, Okio, Retrofit, SQLDelight…) don’t allow contributors to use LLMs to generate code or write PR descriptions/comments anymore.

Can’t blame him. It’s really tiring and risky to keep up with changes submitted by people who don’t fully understand them, especially in an OSS library. It also feels disrespectful to the maintainers, if you ask me.

Is it the best decision for OSS? I’m curious to hear what you think.

https://twitter.com/JorgeCastilloPr/status/2088636197526515715

Posted on

Don’t waste 2 years learning to become an AI agentic engineer in 2026.

https://twitter.com/Dipanshu_AI/status/2083396259747090812

Andrew Ng, the godfather of AI, gave the complete playbook to become one from scratch.

1 hour course. Free:

  • 00:00 – AI agent basics
  • 12:12 – AI Agentic workflows & design patterns
  • 53:27 – Practical tips for building AI agents
  • 1:20:30 – self-improving AI agent loops
  • 1:30:19 – multi-agent AI systems

Posted on

Visa Vulnerability Agentic Harness

Компания Visa выложила в интернет свой код для автономного обнаружения, устранения и проверки уязвимостей с использованием ИИ. Поддерживаются закрытые модели, api и локальные нейросети.

https://twitter.com/MLarchanka/status/2081304015694877014

https://github.com/visa/visa-vulnerability-agentic-harness