learners reached
6M+
based in Sydney
AI courses and live workshops
Learn to build AI systems
from first principles.
Live courses and workshops for engineers, data scientists, analysts, founders, and technical teams building with AI. Learn agents, evals, context engineering, and AI application development through hands-on systems and first-principles instruction.
course graduates
350+
podcast episodes
200+
workshops
Live · hands-on
Featured workshop
next scheduled live session
Courses and workshops
registration, schedule, format, and tuition
Live workshop
Build AI Agents from First Principles
Build the core architecture behind production-ready AI agents: tools, control flow, memory, evaluation, and the practical decisions that separate useful systems from fragile demos.
Registration: open now
Schedule: Jul 29, 2026 PT
Format: 4-hour live workshop
Tuition: USD $500 (early bird $350 until Jul 6; $425 until Jul 19)
View course page →
Live cohort
Build Production-Ready AI Agents for the Enterprise
Build and reason about enterprise agents grounded in your team's knowledge base, with attention to retrieval, evaluation, and production constraints. Co-taught by Hugo Bowne-Anderson and Doug Turnbull, who led search at Reddit and Shopify.
Registration: open now
Schedule: Aug 27-Sep 7, 2026
Format: multi-session live cohort
Tuition: USD $950 (early bird $750 until Jul 6; $850 until Jul 31)
View course page →
PyMC Labs + VG
Master Agentic Data Science
A live workshop for data scientists, analysts, and engineers who want to use AI agents for structured data science without sacrificing statistical rigor or reproducibility. Co-taught with Thomas Wiecki and Luca Fiaschi of PyMC Labs.
Registration: open now
Schedule: Aug 4, 6, 11, 13, 2026
Time: Tue & Thu, 8-11 AM ET / 12:00-15:00 UTC
Format: 12 hours live instruction, 4 sessions
Tuition: USD $1,900
View course page →
Live course
Building AI Applications for Data Scientists and Software Engineers
Build AI-powered software reliably from first principles: agents, evals, iteration, product loops, and the GenAI software development lifecycle. Co-taught by Hugo Bowne-Anderson and Stefan Krawczyk, who works on Agentforce at Salesforce.
Registration: waitlist open
Schedule: final cohort completed; join the waitlist for what comes next
Format: live cohort course
Tuition: USD $2,100
View course page →
Trusted by builders shipping AI systems
engineers from OpenAI, Waymo, Netflix, Meta, Amazon, Maven Clinic, and the Python ecosystem
Shows how solutions are deployed in production.
Applying these ideas to observability and evaluation for our AI agent.
Pragmatic understanding unlocks real-world projects.
Cuts through complexity while staying grounded in practical systems.
From the practitioner ecosystem
courses shaped by real AI, data, and software work
Working patterns from the practitioner ecosystem.
Hugo's courses draw on Vanishing Gradients, High Signal, and years of conversations with people shaping modern AI, data science, Python, and production software.
fast.aiJeremy Howard
StanfordFei-Fei Li
pandas · PositWes McKinney
AnacondaTravis Oliphant · Peter Wang
PydanticSamuel Colvin
O'Reilly · HBRauthor and contributor
For individuals and teams
open cohorts · corporate training · custom team sessions
Open courses
Join a live cohort when you want structured instruction, a hands-on technical curriculum, and a community of peers building with AI.
- Register directly through each course page
- Learn through live sessions, examples, and implementation work
- Best for individuals and small groups joining scheduled cohorts
Corporate training
For companies, teams, and private groups, the material can be adapted to your stack, workflows, and priorities.
- Private cohorts for engineering, data, product, and leadership teams
- Custom emphasis on agents, evals, data science, or AI application development
- Email contact@vanishinggradients.co for availability and scope
Bring the material to your team.
For corporate training, private cohorts, or a workshop adapted to your team’s AI roadmap, send a short note with team size, timeline, and the topic you care about most.
include team size, preferred timing, and the course topic you want to adapt