Jeff Dean

“Also covered by: @Jeff Dean”

#2 𝕏 Google DeepMind launched Gemma 4, a lineup of 7B–196B-parameter foundation models with up to 100K-token contexts and multimodal capabilities. Developers can now access open-source weights, code samples, and tutorials via Vertex AI and GitHub to jumpstart building AI apps. Also covered by: @Jeff Dean

“Jeff Dean asked Gemini to analyze all billboards listed on 101ads.org and generate a report categorizing each company by industry.”

#13 𝕏 Jeff Dean asked Gemini to analyze all billboards listed on 101ads.org and generate a report categorizing each company by industry.

“Jeff Dean asked Gemini to analyze all billboards listed on 101ads.org and generate a report categorizing each company by industry.”

Jeff Dean asked Gemini to analyze all billboards listed on 101ads.org and generate a report categorizing each company by industry. #14 𝕏 Philipp Schmid shared five essential principles from his talk on why senior engineers struggle with AI agents: treating text as state, handing over control, viewing errors as inputs, shifting from unit tests to evals, and designing evolving agents instead of static APIs.

“Also covered by: @Sebastian Raschka , @Simon Willison , @Philipp Schmid , @Jeff Dean , @Google DeepMind , @Demis Hassabis , @Demis Hassabis , @Sebastian Raschka”

Google DeepMind Releases Gemma 4 Open Models #1 𝕏 Google DeepMind launched Gemma 4, a family of Apache 2.0–licensed open models you can run on your own hardware for advanced reasoning and agentic workflows. Also covered by: @Sebastian Raschka , @Simon Willison , @Philipp Schmid , @Jeff Dean , @Google DeepMind , @Demis Hassabis , @Demis Hassabis , @Sebastian Raschka #2 𝕏 Qwen unveiled Qwen3.6-Plus, a next-gen multimodal agentic model with smarter, faster coding execution, sharper vision reasoning and a 1M-token context window by default via API, all while maintaining top-tier general performance.

“𝕏 Jeff Dean unveiled Waxal, a large-scale open resource comprising speech recordings, transcripts, and evaluation tools for dozens of African languages, aiming to accelerate speech-technology research.”

𝕏 Jeff Dean unveiled Waxal, a large-scale open resource comprising speech recordings, transcripts, and evaluation tools for dozens of African languages, aiming to accelerate speech-technology research. #4 𝕏 Andrej Karpathy packaged the “autoresearch” project into a ~630-line, single-GPU repo that runs autonomous 5-minute LLM training loops.

“Also covered by: @Jeff Dean , @Logan Kilpatrick , @Sundar Pichai , @Simon Willison #20 𝕏 Jeff Dean shows that Gemini 3.1 Flash Lite outpaces Gemini 2.5 Flash with much higher tokens/sec throughput and accomplishes complex tasks using only about one-third the tokens.”

Jeff Dean is listed among the people covering Gemini 3.1 Flash-Lite, and later referenced for a throughput comparison post.

“#19 𝕏 Jeff Dean highlights AI’s educational potential and Google’s launch of Gemini training for all 6 million U.S. K–12 and higher-ed teachers, featuring concise, flexible modules with real-world examples and badges to certify AI literacy.”

GenAI PM Daily February 24, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 23 insights for PM Builders, ranked by relevance from Blogs, YouTube, X, and LinkedIn. OpenAI Updates SWE-bench Verified Metrics #1 📝 OpenAI News Why SWE-bench Verified no longer measures frontier coding capabilities - OpenAI explains why the SWE-bench Verified benchmark is no longer used to measure frontier coding capabilities, outlining limitations of the metric and reasons it can misrepresent real-world model performance. The piece describes the rationale for retiring or deprioritizing the benchmark and points toward alternative evaluation approaches for assessing coding ability. Also covered by: @Sebastian Raschka #2 📝 Simon Willison Ladybird adopts Rust, with help by AI - Andreas Kling describes using coding agents (Claude Code and Codex) to port Ladybird's LibJS JavaScript engine from C++ to Rust, producing byte-for-byte identical output and completing ~25,000 lines of Rust in about two weeks.

“Basic Research ROI : Jeff Dean @JeffDean underscored the impact of basic research on innovation in an interview with Magdalena Balazinska, and cited colleague David Patterson’s note that $100 M over 40 years yielded a 1000× return through breakthroughs like RISC and RAID (Patterson’s account) .”

AI Industry Developments & News Basic Research ROI : Jeff Dean @JeffDean underscored the impact of basic research on innovation in an interview with Magdalena Balazinska, and cited colleague David Patterson’s note that $100 M over 40 years yielded a 1000× return through breakthroughs like RISC and RAID (Patterson’s account) . Delethink for Efficient LLM Reasoning : DeepLearningAI @DeepLearningAI presented Delethink , a reinforcement learning approach that trains language models to truncate chains of thought , cutting costs and enhancing long-context reasoning performance.

“Jeff Dean @JeffDean highlighted gathering better and more multilingual training data to improve language & translation models , with TranslateGemma as a downstream result .”

AI Industry Developments & News Open translation models for edge : Demis Hassabis @demishassabis launched TranslateGemma , open translation models built on Gemma 3 for edge devices, outperforming models twice their size across 55 languages . Multilingual training data push : Jeff Dean @JeffDean highlighted gathering better and more multilingual training data to improve language & translation models , with TranslateGemma as a downstream result .

“MedGemma & MedASR release: Jeff Dean @JeffDean released an updated MedGemma model with major accuracy improvements for medical tasks and introduced MedASR for low-error medical speech recognition.”

Google AI Introduces Personal Intelligence From X AI Product Launches & Updates Veo 3.1 improvements: Demis Hassabis @demishassabis announced enhanced expressiveness, portrait mode support, and 1080p & 4K upscaling in Veo 3.1 , rolling out today in the GeminiApp and Flow by Google. Personal Intelligence launch: Google AI @GoogleAI introduced Personal Intelligence in GeminiApp , enabling permissioned reasoning across Gmail , YouTube , Google Photos , and Search for hyper-relevant responses. MedGemma & MedASR release: Jeff Dean @JeffDean released an updated MedGemma model with major accuracy improvements for medical tasks and introduced MedASR for low-error medical speech recognition.