No slides, no hype. I teach by building live on YouTube and X: AI agents, local models, and AI workflows, mistakes included. Join live and ask questions in chat, catch the replays on YouTube, and follow on X for the latest AI updates.
people trained through talks, workshops, and mentoring
2.2k+
GitHub stars on the open-source tools I teach with
25+
years shipping software, the context behind every video
Workshops & talks
Bring an AI workshop to your school or organization.
The same way I teach on the livestreams, run for your students, faculty, or team: hands-on sessions on AI Agents, Local Models, and AI Workflows, in plain language. Online via Zoom, with on-site options on request.
OpenAI said (Nov 3) it is teaching its models to help engineers design, simulate, and build physical products, across mechanical and electrical engineering and end-to-end system design. In an internal test, GPT-6 set the requirements for a device people can use to talk to Codex, and GPT-6 Astra handled the look, enclosure, internal structures, component selection, printed circuit board, interface, firmware, and software connections, with no direct human edits to those design files. Two people did that as a side project in under a week, and a contract manufacturer built a small batch of prototypes in under a month. OpenAI says Astra still missed some production limits, tolerances, and assembly details, so the design still needed review before manufacturing. Since February, token use across OpenAI’s hardware teams has grown 23-fold. On BenchCAD 2.0, which asks models to reconstruct 3D parts and assemblies and identify connections on circuit boards, a newer internal model scored 0.875 on average, compared with 0.594 for GPT-6 Astra at its highest reasoning setting. OpenAI is expanding work with Chip Ganassi Racing to turn 3D scans into editable CAD and 2D drawings when the original files are missing, and it announced a research collaboration with Mazda on airflow and crash-safety simulations. Partner engineers review the outputs. Those scores and the token figure are OpenAI’s. Distinct from the Grid Protectors power program.
Google said (Nov 3) it is adding Guided Learning to Google Search and Google Classroom, and changing how K-12 schools control AI products. Guided Learning coaches a student with questions and step-by-step prompts instead of handing over a final answer. In Search, the change is for student accounts: when Google’s systems read a request as certain kinds of homework help, such as a physics problem or a book report, Search walks through the concepts. Google is starting with a pilot in select U.S. schools later this month. In Google Classroom, Guided Learning is globally available starting today. A teacher assigns an activity anchored in class material, and teachers can see where the class may need more help. Google also says access to Gemini, Gemini Notebook, Gemini in Classroom, and Google Search will be off for K-12 student accounts until a school administrator turns them on. Google says it will tell schools more before that control change rolls out. Distinct from the same-day Tokyo and UK study and from the Google.org teacher-funding announcement.
IBM Research said (Nov 3) it fine-tuned Palette, a 20-billion-parameter open-weight model, to build slide decks in a specific design language. The training data follows the IBM Design Language. IBM splits the job the way a team does: a designer plans the deck and a coder draws each slide, and a harness checks the result. On PresentBench, IBM says Palette ranks first on Economics and Education, third on Talk, and second across the three-domain mean, at 59.7 versus 60.8 for NotebookLM, 1.1 points behind, and ahead of the other systems in that comparison. A separate frontier model drafted each plan from PresentBench’s source document, and Palette built the deck. Across 131 decks, IBM says the harness adds 3.5 points on average. IBM says PresentBench does not measure whether a deck follows a house style. IBM says the baseline numbers come from the PresentBench paper. Those ranks are IBM’s. Distinct from IBM’s same-day CUGA harness comparison.
Source: IBM Research
Updated November 3, 2026
What I teach: AI Agents, Local Models & AI Workflows
AI Agents
Autonomous agents with memory, tools, and skills that do the work, not just chat about it.
I teach AI Agents, Local Models, and AI Workflows to people who want to use AI, not just read about it. Before the channel: senior engineering roles at Standard Chartered Bank and Ohmyhome (Nasdaq), then a run of AI products built and shipped end to end. That's why every video, talk, and post is grounded in what actually ships.
Livestreams AI builds on YouTube and X
Speaker at NTU Singapore, PSIA, and DICT Philippines
300+ trained through workshops and mentoring
Open-source maintainer: Codex orchestrator, Local Evals, Agent Monitor
Subscribe on YouTube for the next livestream and follow on X for the latest AI updates. Want a workshop for your school or organization, or help with a product? Book a call.
