AI/ML × Cloud × DevOps × MLOps

Filesystem, permissions, processes, shell expansion, redirection, pipes, grep, regex, process management, and shell scripting.

Concise, structured companion for fast daily command line lookups and syntax reinforcement.

Complete levels 1 through 25. Forces active practice with SSH, pipes, file permissions, and utilities.

Network layers, IP addressing, CIDR subnets, routing, TCP handshakes, UDP, ports, DNS, NAT, and sockets.

Deeper reference on TCP flow control, congestion avoidance, routing protocols, and transport layer guarantees.

Write one tiny TCP server/client socket program yourself to make low-level socket mechanics concrete.

Main curriculum for Stage 2. Build, run, configure, and orchestrate containerized services with Docker and Docker Compose.

Multi-stage builds, base-image selection, layer caching, .dockerignore, ephemeral containers, and pinned tags.

Understand service discovery, network aliases, container communication, and environment variable interpolation.

Differentiate: image read-only layers vs container writable layer vs bind mounts vs named volume persistence.

Structured book reference to fill conceptual runtime gaps following the Helsinki MOOC.

Self-paced foundational modules: IAM, EC2, VPC, S3, RDS, and CloudWatch metrics.

The 6 pillars: Operational Excellence, Security, Reliability, Performance Efficiency, Cost Optimization, Sustainability.

Examine vetted Reference Architecture Diagrams for web applications and microservices.

Operational reliability, timeouts, retries, jitter, and graceful degradation in production.

Bookmark IAM, VPC, EC2, RDS, S3, ALB, Route 53, and Secrets Manager reference pages.

Triggers, runners, action steps, environment secrets, matrix builds, and deployment environments.

Config in environment, stateless processes, concurrency, disposability, and dev/prod parity.

Pipeline permissions, secret leakage prevention, dependency poisoning, third-party actions, artifact integrity.

Study chapters on automation, release engineering, monitoring, and incident management.

DORA metrics, continuous delivery practices, and high-performance engineering culture.

Comprehensive principles on automated acceptance testing, release pipelines, and deployment strategies.

Step-by-step tutorial on provisioning and managing AWS resources via Terraform.

Remote backends, state locking, state inspection, resource drift, and disaster recovery.

Build and publish reusable, composable modules for networking, compute, and data layers.

Write automated unit and integration tests for Terraform code with ephemeral infrastructure and mocks.

Dedicated guides for ALBs, Auto Scaling Groups, RDS instances, and IAM permission policies.

Production architectures, multi-account layouts, and operational best practices.

Primary ML book. Study Ch. 2, 3, 4, 5, 6, 8, 9, 12. Complete the Python labs.

Regression, classification, loss curves, data representations, and generalization.

Practical tabular preprocessing, ColumnTransformers, cross-validation, and hyperparameter tuning.

Targeted reference for linear algebra, vector calculus, probability, and optimization theory.

Industry standard practical reference for building production ML pipelines.

Tensors, Datasets & DataLoaders, Transforms, Neural Network modules, Autograd, and Optimization loops.

The prioritized deep learning textbook: loss functions, optimization, regularization, CNNs, and transformers.

Interactive textbook with complete PyTorch code implementations for every deep learning architecture.

Celebrated course notes on gradient mechanics, backpropagation intuition, and neural net debugging.

Transformers, Tokenizers, Datasets, and fine-tuning. Study after understanding PyTorch training loops.

Read all 43 rules on pipeline discipline, simple baselines, and avoiding training-serving skew.

Seminal paper on glue code, pipeline jungles, dead code paths, and configuration debt.

End-to-end design, data tracking, model serving, testing, and reproducibility.

Track runs, metrics, parameters, and deploy a self-hosted tracking server backed by S3 and Postgres.

Model versioning, lineage tags, champion/challenger aliases, and deployment lifecycle states.

Notes and labs on ML infrastructure, experiment management, and testing ML pipelines.

Holistic production engineering guide connecting data pipelines to inference architectures.

Vendor-neutral conceptual standard for MLOps maturity levels: manual, CI/CD, and CI/CD/CT pipelines.

Versioning, testing, automation, reproducibility, and deployment governance for ML systems.

Project instructions and code covering experiment tracking, orchestration, deployment, and monitoring.

Production jobs, services, automated testing pipelines, and data engineering patterns.

Continual learning strategies, deployment architectures, and automated model monitoring.

Pods, Deployments, Services, ConfigMaps, Secrets, Volumes, Probes, Requests/Limits, Namespaces.

Main Kubernetes practice curriculum. Multi-tier application management on real clusters.

Disposable browser clusters for rapid debugging and cluster failure experiments.

Chart.yaml, values.yaml, template pipelines, functions, conditionals, and debugging.

IRSA, VPC CNI, Karpenter autoscaling, security, and AWS observability.

High-level architectural reference written by the creators of Kubernetes.

Structured logs, dimensional metrics, distributed tracing, and spans.

Metrics scraping, PromQL queries, counters, histograms, and alerting rules.

Production dashboard construction, alert routing, and metrics visualization.

Distributed monitoring, SLI/SLO design, alerting thresholds, and blameless postmortems.

All 40 text lessons: data quality, batch vs real-time monitoring, model quality, and drift.

Crucial principle: understand when data drift does and does not degrade model performance.

The definitive guide for Linux CPU, memory, disk, network, and latency profiling.

Practical guide to defining meaningful SLIs and managing error budgets.

High-cardinality telemetry, distributed tracing, and debugging unknown-unknown failures.

Audit your completed platform against the 6 pillars before finalizing documentation.

Structured review checklist specifically for ML data, training, deployment, and monitoring workloads.

Provide System Context, Container, and Deployment architecture diagrams in your final portfolio.

Security audit checklist covering authentication, secrets, endpoints, and container isolation.

Verify timeouts, backoff, idempotency, and overload prevention across all services.