Learn
Structured learning on LLM inference, ML infrastructure, and production AI — taught from 15+ years of real engineering experience.
LLM Inference Systems — Deep Dive
8-week live cohort course
The most comprehensive course on LLM inference available anywhere. Built from production experience, not theory. You'll go from understanding the basics of serving to designing multi-node distributed inference systems.
What's Covered:
- LLM serving fundamentals — request lifecycle, tokenization, decoding
- KV cache internals — memory layout, eviction policies, prefix caching
- Continuous batching and PagedAttention deep dive
- Tensor parallelism and pipeline parallelism
- Speculative decoding and model quantization
- vLLM architecture and deployment patterns
- Multi-node distributed inference at scale
- Cost modeling, capacity planning, and SLA design
LLM Inference Systems — Self-Paced
Recorded version of the live cohort
The full live cohort recorded and structured for self-paced learning. Same curriculum, same depth — learn on your own schedule. Includes all materials, code examples, and architecture diagrams.
What's Covered:
- Complete recorded sessions from the live cohort
- Architecture diagrams and reference materials
- Code walkthroughs and implementation examples
- Self-assessment quizzes per module
- Community access for questions
- Free updates when content is refreshed
ML Systems Mentorship Cohort
Small-group mentorship program
A small-group (8–12 people) mentorship program for senior ML engineers and engineering managers looking to level up on AI systems architecture and LLM infrastructure. Weekly group sessions plus 1:1 time.
What's Covered:
- Weekly group sessions (90 min) on ML systems topics
- 2× 1:1 sessions per month (30 min each)
- Personalized learning path and reading list
- Career guidance for senior/staff/principal roles
- Architecture review of your current system
- Access to all recorded course content
Corporate Training & Workshops
Custom training for teams building or scaling AI systems. I design workshops around your specific stack, use cases, and skill gaps — from half-day intensives to multi-week deep dives.
- Custom curriculum tailored to your team's stack
- On-site or virtual delivery
- Workshops from half-day to multi-week engagements
- Architecture review + team upskilling combined
- Post-workshop Q&A and follow-up sessions
Be First to Know
All courses launch to Substack subscribers first — with early-bird pricing and preview access.
Subscribe to Stay Updated