AI researcher and educator
Andrej Karpathy
Andrej Karpathy is an AI researcher and educator, a founding member of OpenAI, and a former director of AI at Tesla working on computer vision.
About Andrej Karpathy
Andrej Karpathy is an AI researcher and educator whose public teaching covers neural networks and large language models. His educational videos include both technical lessons on building models and introductions for a general audience.
He was a founding member and research scientist at OpenAI, led Tesla's AI and computer vision work from 2017 to 2022, and returned to OpenAI in 2023 to work on midtraining and synthetic data. During his PhD at Stanford, he designed and taught the CS231n course on deep learning for visual recognition.
Mentioned in our coverage
117 articles name Andrej Karpathy, excluding author credits.
Page 1 of 10
- Australian STEM Videogame Challenge Ends After AI Game Submissions →
The Australian STEM Videogame Challenge will hold its final competition in 2026 after ACER — which runs the national schools program — said generative AI made it impossible to fairly judge which parts of submitted games students built themselves. PC Gamer confirmed the decision with ACER leadership on November 10–11, 2026, after judges on Reddit raised the same concern.
- WikiTrace: Evidence-Linked Agent Memory When Lessons Hurt Performance →
Agent memory can make models fail tasks they would solve without memory — KAIST AI calls that memory-induced degradation. WikiTrace links each lesson to the run it came from so agents descend into evidence before acting. Code is on GitHub under Apache-2.0.
- OpenAI: What Changes Chain-of-Thought Monitorability Across Training →
On November 10, 2026 OpenAI Alignment published a lifecycle study of chain-of-thought monitorability — how better base models, RL, architecture choices, and teacher rollouts affect no-CoT reasoning and CoT evasion. The post ties directly to GPT-6 Astra deployment vetting and UK AISI no-CoT reviews.
- Top 20 Things Built With Claude Fable 5.2 (Shumer, Anthropic, Mollick) →
Claude Fable 5.2 shipped on November 9, 2026. Within a day, Matt Shumer posted six demos from 400B tokens of testing, Anthropic published an OS built by 1,000 subagents, and Ethan Mollick shipped a world-building site. Here are 20 builds, ranked, with what is verified and what is a creator claim.
- Jensen Huang Cancels ElevenLabs Summit and Anthropic CEO Forum →
On November 9, 2026 TechCrunch reported that Jensen Huang withdrew from ElevenLabs's New York summit and an Anthropic CEO Forum. ElevenLabs confirmed the change. The timing follows Huang's apology memo and Nvidia's board investigation into reports of foreign intelligence contact.