HeyJaredHands On AI Agent Mastery Course
Learning Outcomes By course completion, you'll have: Built 15+ production-ready agents with proper error handling and monitoring Implemented scalable architectures handling concurrent requests and resource management Deployed agent systems with CI/CD pip
- Indexed issues, last 90 days
- 15
- Latest publication
- Sep 29, 2026
- Audience
- Checking…
- Earliest in this view
- Aug 18, 2026
Latest issues
Lesson 24: SOC 2 Compliance — Audit Trail & Access Controls (opens the original)
Read excerpt
IntroductionEnterprise AI agents handle sensitive data, invoke external tools, and incur measurable cost. Auditors expect proof that access is controlled, actions are logged immutably, and tampering is detectable. This lesson implements SOC 2–aligned audit trails with chain-hash integrity, RBAC evidence, and an auditor-ready evidence pack. You simulate immutable S3 writes, verify tamper detection on a linked hash chain, and validate seven production controls on a live FastAPI dashboard.Highlight
Lesson 22: Kubernetes Deployment — Production Manifests (opens the original)
Read excerpt
IntroductionShipping an AI agent to production is not a Docker tag alone. Clusters need Deployment, Service, HPA, ConfigMap, and Secrets that encode resource bounds, probes, and zero-downtime rollouts. This lesson turns those manifests into a checkable contract: validate rolling-update strategy, readiness and liveness paths, CPU/memory limits, and the rule that secrets never land in ConfigMaps. A FastAPI dashboard and CLI demo simulate checklist passes and rolling-update waves without requiring
Lesson 23: Auto-Scaling — KEDA & Custom Metrics (opens the original)
Read excerpt
IntroductionCPU-based autoscaling lags behind AI agent workloads: spikes arrive when requests queue, not after inference finishes. This lesson replaces reactive CPU scaling with KEDA-driven scale on agent_request_queue_depth, exposed through the Kubernetes custom metrics API via a Prometheus adapter. You validate ScaledObject triggers, simulate burst scale-up and scale-to-zero, and quantify FinOps savings on a live dashboard.HighlightsKEDA ScaledObject: Prometheus queue-depth trigger plus cron b
Lesson 21: The Full LLMOps Pipeline — Closing the Loop (opens the original)
Read excerpt
IntroductionProduction agents do not improve because someone edits a prompt in a notebook once. They improve when traces from real traffic become evaluation cases, candidates are measured against a baseline, winners pass a statistical gate, and deployment is auditable. This lesson implements that closed loop with anonymised trace ingestion, a deterministic eval harness, A/B selection, and a deploy step gated on a minimum improvement threshold. A FastAPI dashboard exposes every stage so operators
Lesson 20: Knowledge Graph Integration — Structured Retrieval (opens the original)
Read excerpt
IntroductionEmbeddings retrieve similar text, but production agents also need exact relationship answers: which tools a user triggered, which agents called them, and what each call cost. This lesson builds an operational knowledge graph for Agents, Tools, Users, Sessions, and AuditEvents. An in-memory backend powers demos and tests; Neo4j is optional for production-shaped deployments. A FastAPI dashboard proves every counter moves after demo traffic so structured retrieval stays observable, not
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.