Digital toolkits, architecture guides, and async cohorts built directly from the AIEWorks newsletter. One purchase. Lifetime access. No live sessions required.
Every Concept Note and the complete Arc 1 build-along are permanently free. Subscribe once and these never move behind a paywall.
Why every AI support ticket is already a perfectly structured eval case — and the habit that closes the gap.
Read on Substack →Build the eval case dataset format, versioning schema, and the CLI that turns tickets into test cases. Runnable repo included.
Read on Substack →The structural gap between traditional test planning and AI feature testing — for QA engineers and engineering managers.
Read on Substack →The original 60-day hands-on course. Days 1–3 unlocked. System architecture, requirements, database schema — all free.
View curriculum →180-day curriculum from Python fundamentals to deep learning. Foundation for everything in the vault.
View curriculum →What an AI system design review must cover that a traditional architecture review does not.
Read on Substack →Toolkits, playbooks, and reference packs extracted directly from the newsletter's build-along arcs. Buy once, use forever.
Everything from Arc 1 in one download. Golden dataset schema templates, 50 pre-written eval case starters for common AI failure modes, CI gate GitHub Actions workflow, and the Streamlit PM dashboard template.
The annotated guide to building eval systems that don't lie to you. Covers LLM-as-judge calibration, inter-rater agreement, and regression testing for non-deterministic systems.
Checklists and templates for every role on the team. The PM writes a better spec. The designer handles model uncertainty. The SRE knows what to monitor. Built for teams, not just engineers.
The exact production system used to write the newsletter. Full prompt, banned-phrase automation, post checklist, SVG visual language guide, and Substack scheduling templates.
Every architecture, data flow, state machine, and sequence diagram from all four production arcs. Editable SVG source files with the full color system guide.
Structured learning built around the newsletter's build-along arcs. Weekly exercises, peer community, recorded Q&As, and a completion certificate. Zero scheduling. 100% async.
Follow Arc 1 as a structured program. Each week adds one layer to the eval harness — golden dataset, automated scorer, LLM-as-judge, CI gate. Ship a complete system in four weeks.
All four production arcs as a structured 13-week program. Go from eval fundamentals to a complete AI reliability stack: proof system, feature flags, audit trail, spend governance. Portfolio-grade output.
Designed for delivery pods and engineering teams, not solo learners. Two arcs, structured for multi-role participation. The PM writes the spec, the engineer builds, the QA evaluates, the SRE deploys. Everyone ships the same system from their own role.
Everything paired the way it gets used. One price, one Drive folder, one Discord invite.
Arc 1 Cohort (4 weeks) + AI Feature Proof System Toolkit. Everything you need to build and ship your first eval system.
Full Stack Cohort (13 weeks) + all 5 digital products. The complete production AI engineering stack — cohort, toolkits, playbook, role pack, diagram pack, and prompt kit.
Cross-Functional Team Cohort (8 weeks, 5 seats) + AI-Native SDLC Role Pack (5 copies). Ideal for IT services delivery pods, engineering squads, and tech leads running AI transformation projects.