AI Specialization Program
•
Soft AI Academy
AI Specialization Program
•
Technical Tracks
AI Engineering & Data
•
Week 6
WEEK 6:
Computer Vision + Deployment + AI Systems Engineering + Automation Architecture
Theme: Deploy & Scale AI Systems
Week 6 shifts from modeling → real-world system deployment.
Week 6 —
6 Lesson Structure
Lesson 1:
Computer Vision Fundamentals
Focus:
Images as numerical arrays
CNN intuition
Image classification
Object detection basics
Real-world CV applications
Outcome: Students understand how machines interpret images.
Lesson 2:
Building Basic Vision Models
Focus:
Using PyTorch for image models
Pre-trained vision models
Transfer learning concept
Evaluating CV systems
Outcome: Students can implement a simple vision pipeline.
Lesson 3:
Deployment Basics (FastAPI / Streamlit)
Focus:
What is model serving?
API-based deployment
Simple FastAPI app
Streamlit demo apps
Turning models into usable tools
AI Systems Engineering fundamentals
Service-oriented AI architecture
AI components and system boundaries
Designing reusable AI services
Scaling AI beyond a single model
Cloud deployment awareness
Outcome:
Students understand how deployed models fit into larger AI systems. They can expose models via APIs or simple interfaces.
Lesson 4:
Docker & Reproducible Environments
Focus:
What is containerization?
Why Docker matters
Basic Dockerfile
Running model inside container
Environment reproducibility
Production Pipeline Design
Development → Testing → Deployment flow
CI/CD concepts for AI systems
Reproducible production environments
Deployment consistency principles
Outcome:
Students understand portable AI deployment and how AI systems move from development into production reliably.
Lesson 5:
Monitoring, Guardrails & Drift
Focus:
Model monitoring basics
Performance tracking
Data drift detection
Guardrails for GenAI
Logging & alerting mindset
Automation Architecture:
Designing resilient AI workflows
Failure handling strategies
Monitoring automated systems
Safety checkpoints
Human approval layers
Outcome:
Students understand post-deployment responsibilities and how to monitor and manage automated AI systems safely.
Lesson 6:
AI Automation Pipelines (Capstone Integration)
Focus:
Automating workflows
Connecting APIs
LLM + CV integration possibilities
Designing production pipelines
AI Workflow Orchestration:
Coordinating multiple AI services
Workflow management concepts
Sequential vs parallel workflows
Multi-Step AI Systems:
Input → Processing → Decision → Action
Multi-stage AI pipelines
Agent Workflows;:
Planning
Tool execution
Memory concepts
Workflow coordination
End-to-End Automation Design:
AI + APIs + Databases
AI + Vision + LLM systems
Enterprise automation thinking
Capstone preparation guidance
Outcome
Students can design orchestrated, multi-step AI systems that combine models, tools, APIs, and workflows into production-ready solutions.
They can design a full AI system architecture.
Week 6 Final Outcome
By the end of Week 6, students can:
✔ Build basic computer vision systems
✔ Deploy ML or LLM models
✔ Create API-based AI services
✔ Understand containerization
✔ Monitor deployed systems
✔ Design automation pipelines
✔ Think end-to-end AI architecture
Week 6 produces:
Deployment-ready AI system engineers.
Full 6-Week Technical Track Summary
Week 1 → Engineering Foundations
Week 2 → Data Engineering
Week 3 → Machine Learning
Week 4 → Deep Learning + MLOps
Week 5 → NLP & Generative AI
Week 6 → Computer Vision + Deployment
This is now a complete, production-grade AI Engineering specialization.
As we conclude week 6 and move on to weeks 7 and finally week 8.
Powered by Soft AI Africa | Training the Next Generation of AI Leaders in Africa.