←All Series
🧭
Frontier Model Engineering
Four hands-on workshops for taking any model release from artifact audit to durable agents, long-context storage, and topology-aware multi-GPU inference.
complete4 / 4 episodes•74 min total•advanced
Series Progress100%
What You'll Learn
- ✓Audit a model release without executing unreviewed repository code
- ✓Design agent side effects around durable intent and recovery
- ✓Separate configured, resident, retrieved, cached, and useful context
- ✓Build per-rank GPU placement and communication plans
- ✓Publish evidence and workshop artifacts another engineer can reproduce
Episodes by Track
🔎
Model Forensics
Inspect identities, artifacts, tensors, code, licenses, and evaluation claims before loading a model.
1 post
🧰
Runtime Systems
Treat agent state and long context as durable protocol and storage problems.
2 posts
🖧
Inference Infrastructure
Place weights and cache across GPUs, then map communication onto real topology.
1 post
Prerequisites
- •Python and command-line tooling
- •Transformer fundamentals
- •Distributed-systems basics
- •Comfort reading model configs and API schemas
Who This Is For
- •ml-engineers
- •platform-engineers
- •inference-engineers
- •ai-developers