Skip to content

PREREQUISITES.md

Latest commit

 

History

History
554 lines (401 loc) · 21.2 KB

PREREQUISITES.md

File metadata and controls

554 lines (401 loc) · 21.2 KB
Raw

Prerequisites for AI Infrastructure Engineer Curriculum

This curriculum is designed for engineers looking to advance their AI infrastructure skills to a mid-level professional standard. Before starting, ensure you have the necessary foundation.

🎯 Quick Assessment

Already completed the Junior AI Infrastructure Engineer curriculum? ✅ You're ready to start!

Haven't completed Junior curriculum? → Use the self-assessment below to check your readiness.

Option 1: Complete Junior Curriculum (RECOMMENDED)

The Junior AI Infrastructure Engineer curriculum provides ALL prerequisites for this Engineer-level course.

Direct Module Mapping

Every prerequisite for this Engineer curriculum is covered in the Junior curriculum:

Engineer Prerequisite Junior Module Hours What You'll Learn
Python 3.9+ (intermediate level) Module 001: Python Fundamentals 15h OOP, type hints, testing, error handling, decorators, context managers
Linux/Unix command line basics Module 002: Linux Essentials 15h CLI navigation, bash scripting, processes, SSH, networking, system administration
Git fundamentals Module 003: Git & Version Control 10h Branching, merging, rebasing, collaboration workflows, GitHub/GitLab
Basic ML concepts Module 004: ML Basics 20h PyTorch, TensorFlow, training, inference, evaluation, model export
Docker basics Module 005: Docker & Containerization 15h Docker images, containers, multi-stage builds, Docker Compose
Kubernetes intro Module 006: Kubernetes Introduction 20h Pods, deployments, services, ConfigMaps, Secrets, basic operations
API development Module 007: APIs & Web Services 15h REST APIs, FastAPI, Flask, authentication, testing
Database basics Module 008: Databases & SQL 15h SQL fundamentals, PostgreSQL, Redis, data modeling
Monitoring basics Module 009: Monitoring & Logging 15h Prometheus, Grafana, logging, alerting basics
Cloud platforms Module 010: Cloud Platforms 20h AWS, GCP, Azure basics - VMs, storage, networking

Total Junior Curriculum: 440 hours (22 weeks part-time, 11 weeks full-time)

Why Complete Junior First?

  1. Comprehensive Coverage: All prerequisites covered with hands-on practice
  2. Portfolio Building: 5 projects for your resume
  3. Job-Ready Skills: Qualifies you for entry-level AI infrastructure roles
  4. Smooth Transition: Designed to prepare you specifically for this Engineer curriculum
  5. No Gaps: Ensures you have ALL foundational knowledge

Getting Started with Junior Curriculum

# Clone the Junior repository
git clone https://github.com/ai-infra-curriculum/ai-infra-junior-engineer-learning.git
cd ai-infra-junior-engineer-learning

# Follow the getting started guide
cat GETTING_STARTED.md

Option 2: Self-Assessment Checklist

If you haven't completed the Junior curriculum, verify you can perform ALL tasks below. Missing ANY skill? → Complete the relevant Junior module first.

Python Skills ✅

Required for: All Engineer modules

  • Write production-quality Python with OOP, type hints, and comprehensive docstrings
  • Use async/await for concurrent programming
  • Write comprehensive unit tests with pytest (80%+ coverage)
  • Use virtual environments (venv, conda) and dependency management
  • Understand and use decorators, context managers, generators
  • Profile and optimize Python code for performance
  • Handle errors gracefully with custom exceptions
  • Use Python dataclasses and Pydantic for validation

Test Yourself:

# Can you build a production-quality REST API with:
# - Type hints throughout
# - Async endpoints
# - Error handling
# - Comprehensive tests
# - Pydantic validation?

Missing skills? → Junior Module 001

Linux/Unix Skills ✅

Required for: Cloud Computing, Containerization, Kubernetes, all deployment modules

  • Navigate Linux CLI efficiently (cd, ls, grep, sed, awk, find)
  • Write bash scripts for automation (loops, conditionals, functions)
  • Manage processes (ps, top, htop, kill, systemctl)
  • Understand file permissions and ownership (chmod, chown)
  • Configure SSH keys and secure remote access
  • Debug system issues using logs (/var/log, journalctl)
  • Use package managers (apt, yum, brew)
  • Understand environment variables and PATH

Test Yourself:

# Can you write a bash script that:
# - Monitors system resources
# - Logs to syslog
# - Handles errors gracefully
# - Runs as a systemd service?

Missing skills? → Junior Module 002

Git & Version Control Skills ✅

Required for: All modules (especially IaC, MLOps, CI/CD)

  • Use Git workflows (feature branches, pull requests)
  • Perform branching, merging, and rebasing confidently
  • Resolve merge conflicts
  • Understand Git internals (commits, trees, blobs, refs)
  • Use GitHub/GitLab for collaboration and code review
  • Follow conventional commits and semantic versioning
  • Understand GitOps principles
  • Use git hooks for automation

Test Yourself:

# Can you:
# - Set up a GitOps workflow with ArgoCD?
# - Implement pre-commit hooks?
# - Manage multiple release branches?

Missing skills? → Junior Module 003

Machine Learning Skills ✅

Required for: All modules (foundation for all AI infrastructure work)

  • Train basic models with PyTorch or TensorFlow
  • Understand training, validation, and testing splits
  • Perform model evaluation (accuracy, precision, recall, F1, AUC)
  • Export and load trained models (checkpoint, ONNX)
  • Understand overfitting, underfitting, and regularization
  • Use data loaders and preprocessing pipelines
  • Debug training issues (loss not decreasing, NaN values)
  • Understand basic neural network architectures

Test Yourself:

# Can you:
# - Train a CNN for image classification?
# - Implement custom loss functions?
# - Use transfer learning?
# - Export model for production serving?

Missing skills? → Junior Module 004

Docker & Containerization Skills ✅

Required for: Containerization, Kubernetes, MLOps, all deployment modules

  • Build Docker images with Dockerfiles
  • Use multi-stage builds for optimization
  • Run containers with volume mounts and port mapping
  • Use Docker Compose for multi-service applications
  • Understand container networking (bridge, host, overlay)
  • Manage container resources (CPU, memory limits)
  • Debug container issues (logs, exec, inspect)
  • Understand container security basics (user namespaces, read-only filesystems)

Test Yourself:

# Can you:
# - Build optimized ML serving container (<500MB)?
# - Set up multi-container app with Docker Compose?
# - Implement health checks and auto-restart?

Missing skills? → Junior Module 005

Kubernetes Skills ✅

Required for: Kubernetes, MLOps, Monitoring modules

  • Deploy applications to Kubernetes
  • Understand pods, deployments, services, ingress
  • Use ConfigMaps and Secrets for configuration
  • Debug pod issues (logs, describe, exec, port-forward)
  • Understand basic resource management (requests, limits)
  • Use kubectl effectively
  • Understand Kubernetes networking basics
  • Deploy multi-tier applications

Test Yourself:

# Can you:
# - Deploy ML model API to Kubernetes?
# - Set up horizontal pod autoscaling?
# - Configure ingress for external access?
# - Debug pod startup failures?

Missing skills? → Junior Module 006

API Development Skills ✅

Required for: Foundations, MLOps, LLM Infrastructure modules

  • Build REST APIs with FastAPI or Flask
  • Implement authentication and authorization
  • Write API tests (unit and integration)
  • Generate OpenAPI/Swagger documentation
  • Implement rate limiting and caching
  • Handle errors and validation properly
  • Use async endpoints for high concurrency
  • Deploy APIs to production

Test Yourself:

# Can you build an API that:
# - Serves ML model predictions?
# - Has authentication (JWT)?
# - Includes comprehensive tests?
# - Has auto-generated docs?

Missing skills? → Junior Module 007

Database Skills ✅

Required for: Data Pipelines, MLOps modules

  • Write complex SQL queries (joins, aggregations, subqueries)
  • Design normalized database schemas
  • Use PostgreSQL or MySQL for production
  • Understand indexing and query optimization
  • Use Redis for caching
  • Perform database migrations
  • Implement connection pooling
  • Understand ACID properties and transactions

Test Yourself:

-- Can you:
-- - Design schema for ML experiment tracking?
-- - Optimize slow queries?
-- - Implement caching strategy?

Missing skills? → Junior Module 008

Monitoring & Logging Skills ✅

Required for: Monitoring & Observability module

  • Set up Prometheus for metrics collection
  • Create Grafana dashboards
  • Implement application logging (structured logs)
  • Configure alerts based on metrics
  • Understand metrics types (counter, gauge, histogram)
  • Debug issues using logs and metrics
  • Implement log aggregation
  • Understand basic observability concepts

Test Yourself:

# Can you:
# - Set up Prometheus + Grafana for ML API?
# - Create custom metrics for model performance?
# - Configure alerts for service downtime?

Missing skills? → Junior Module 009

Cloud Platform Skills ✅

Required for: Cloud Computing, all deployment modules

  • Deploy VMs and storage on AWS/GCP/Azure
  • Use cloud CLI tools (aws-cli, gcloud, az)
  • Understand cloud networking (VPCs, subnets, security groups)
  • Deploy containerized applications to cloud
  • Monitor cloud resources and costs
  • Use managed services (RDS, Cloud SQL, S3, GCS)
  • Implement basic IAM (users, roles, policies)
  • Understand cloud pricing models

Test Yourself:

# Can you:
# - Deploy Kubernetes cluster to cloud?
# - Set up VPC with public/private subnets?
# - Implement cost monitoring and alerts?

Missing skills? → Junior Module 010

Option 3: Alternative Pathways

If You Have Industry Experience

Engineers with 1-2 years of experience in related fields may be able to skip some Junior modules:

DevOps/SRE Background

Can likely skip: Junior Modules 002, 003, 005 (Linux, Git, Docker) Should review: Modules 004, 006, 009 (ML Basics, Kubernetes, Monitoring) Must complete: Modules 001, 007, 008 (Python for ML, APIs, Databases)

Backend Development Background

Can likely skip: Junior Modules 001, 003, 007, 008 (Python, Git, APIs, Databases) Should review: Modules 005, 006, 009 (Docker, Kubernetes, Monitoring) Must complete: Modules 002, 004 (Linux, ML Basics)

Data Engineering Background

Can likely skip: Junior Modules 001, 008, 009 (Python, Databases, Monitoring) Should review: Modules 002, 003, 005 (Linux, Git, Docker) Must complete: Modules 004, 006 (ML Basics, Kubernetes)

ML/Data Science Background

Can likely skip: Junior Module 004 (ML Basics) Should review: Modules 001, 007 (Python for infrastructure, APIs) Must complete: Modules 002, 003, 005, 006, 009 (Linux, Git, Docker, Kubernetes, Monitoring)

Recommendation: Even with experience, review ALL Junior modules to ensure no gaps. The Junior curriculum focuses on AI/ML infrastructure specifically, which may differ from general software engineering.

Recommended Study Plans

Plan A: Fast Track (Have Most Prerequisites)

Duration: 4-6 weeks

Week 1: Self-assess using checklists above Week 2: Complete any missing Junior modules Week 3: Review Junior projects for hands-on practice Week 4: Quick refresher on weak areas Week 5-6: Begin Engineer Module 101 (Foundations)

Best for: Engineers with 1-2 years DevOps/ML experience

Plan B: Standard Track (Complete Junior First)

Duration: 22-24 weeks

Weeks 1-22: Complete entire Junior curriculum (440 hours) Week 23: Build capstone project and polish portfolio Week 24: Begin Engineer Module 101 (Foundations)

Best for: Career changers, bootcamp grads, recent CS graduates

Plan C: Part-Time Track (Working Professionals)

Duration: 44-48 weeks

Weeks 1-44: Complete Junior curriculum part-time (10 hours/week) Weeks 45-47: Build capstone project Week 48: Begin Engineer curriculum

Best for: Working professionals learning in evenings/weekends

Automated Skill Assessment

Run the Self-Assessment Tool

# Clone Junior repository
git clone https://github.com/ai-infra-curriculum/ai-infra-junior-engineer-learning.git
cd ai-infra-junior-engineer-learning

# Run automated skill assessment
python scripts/assess_skills.py

# Output will show:
# - Current skill levels for each area
# - Recommended modules to complete
# - Estimated time to readiness

The assessment tool checks:

  • Python coding ability (via coding challenges)
  • Linux/Git knowledge (via command simulations)
  • Docker/Kubernetes hands-on (via practical tasks)
  • ML fundamentals (via conceptual questions)
  • Cloud platform familiarity (via scenario questions)

Passing score: 70%+ in all areas Recommended: 85%+ for smooth transition to Engineer curriculum

Getting Help

If You're Unsure About Your Readiness

Option 1: Post in GitHub Discussions

  • Share your background and experience
  • Get personalized recommendations from community

Option 2: Join our Discord Server

  • Ask questions in #prerequisites channel
  • Get feedback from instructors and peers

Option 3: Schedule Office Hours

  • Weekly Q&A sessions (Fridays 3-4pm PT)
  • 1-on-1 prerequisite consultations (by appointment)

Option 4: Review Junior Curriculum Structure

  • Browse the Junior course structure
  • Try the first exercise of each module
  • If you can complete them easily, you may be ready

Success Metrics: Are You Ready?

You're ready for the Engineer curriculum when you can confidently:

Technical Capabilities

  • Deploy a complete ML application (API + Docker + K8s + monitoring) from scratch
  • Debug production issues in distributed systems
  • Write infrastructure automation scripts (Python + Bash)
  • Explain ML model serving architecture and trade-offs
  • Navigate and troubleshoot cloud environments
  • Understand and optimize cloud costs
  • Implement CI/CD pipelines for ML applications
  • Monitor and alert on system and model performance

Assessment Scores

  • Pass Junior curriculum assessments with 80%+ average
  • Complete at least 2 Junior projects with "Excellent" grade
  • Demonstrate proficiency in self-assessment checklist
  • Build a portfolio project demonstrating all prerequisite skills

Time Investment Ready

  • Can commit 40+ hours/week (full-time) or 20+ hours/week (part-time)
  • Have 6+ months available for dedicated learning
  • Can access cloud accounts (AWS/GCP/Azure free tiers)
  • Have suitable hardware (8GB+ RAM, decent CPU, GPU optional)

What Happens If You Start Without Prerequisites?

Scenario 1: Missing Python/Linux/Git basics ❌ Result: Will struggle in Module 101, unable to complete exercises ✅ Fix: Complete Junior Modules 001-003 first (40 hours)

Scenario 2: Missing ML fundamentals ❌ Result: Won't understand ML infrastructure trade-offs ✅ Fix: Complete Junior Module 004 (20 hours)

Scenario 3: Missing Docker/Kubernetes ❌ Result: Can't complete containerization and orchestration modules ✅ Fix: Complete Junior Modules 005-006 (35 hours)

Scenario 4: Missing monitoring/cloud basics ❌ Result: Struggle with observability and deployment modules ✅ Fix: Complete Junior Modules 009-010 (35 hours)

Bottom line: Starting without prerequisites will significantly slow your progress and reduce learning effectiveness. We strongly recommend completing the Junior curriculum or validating all prerequisite skills first.

Prerequisites Timeline Summary

Starting Point Time to Ready Recommended Path
Complete beginner 22-44 weeks Complete full Junior curriculum
Some coding experience 16-22 weeks Complete Junior, skip familiar modules
1-2 years DevOps/Backend 8-12 weeks Self-assess, fill gaps, review modules
1-2 years ML Engineering 6-10 weeks Focus on infrastructure modules (002, 005, 006, 009)
Current Junior grad 0 weeks Start immediately! ✅

Ready to Begin?

✅ If You Have All Prerequisites

Congratulations! You're ready to start the Engineer curriculum.

Next steps:

  1. Read the Getting Started Guide
  2. Set up your development environment
  3. Begin Module 101: Foundations

⚠️ If You're Missing Prerequisites

Don't worry! The Junior curriculum is designed to get you ready.

Next steps:

  1. Visit the Junior AI Infrastructure Engineer repository
  2. Complete relevant modules (or all modules for comprehensive preparation)
  3. Build your portfolio with Junior projects
  4. Return here when ready

Questions?

Can I start the Engineer curriculum while completing Junior prerequisites? Not recommended. You'll be more successful completing prerequisites first.

How long does the Junior curriculum take?

  • Full-time: 11 weeks (40 hours/week)
  • Part-time: 22 weeks (20 hours/week)
  • Working professionals: 44 weeks (10 hours/week)

Do I need to complete ALL Junior modules? If you can pass the self-assessment with 80%+ in all areas, you can skip modules. But we recommend completing all for comprehensive preparation.

Can I get credit for industry experience? The self-assessment checklist allows you to validate skills from any source. If you can demonstrate proficiency, prerequisites are met regardless of where you learned them.

What if I fail the self-assessment? Complete the relevant Junior modules, then retake the assessment. There's no penalty for retaking.

Start Your Journey Today:

👉 Junior AI Infrastructure Engineer Learning Path

👉 Engineer Curriculum Getting Started (if prerequisites complete)

Last Updated: October 2025 Maintained by: AI Infrastructure Curriculum Team