Critical Thinking for AI
Evaluate AI output with confidence. Detect subtle errors, reason through dilemmas, and make sound decisions when AI gives you the wrong answer.
Don't Get Replaced by AI — Learn and Build It
Pakistan's largest hands-on AI Engineering Program
Zero Coding Experience needed. Build AI agents that reason, plan, and act autonomously — then deploy them at production scale.
- Karachi
- Lahore
- Islamabad
- Peshawar
- Faisalabad
100k+
Alumni
5
Quarters
4
Certifications
11
Exams
Presidential AI Initiative
Dr. Arif Alvi
"Artificial Intelligence is the future. Pakistan must equip its youth with cutting-edge AI skills to lead in the global digital economy."
AGENTIC AI PROFESSIONAL TRACK
Five Quarters to Certified Agentic AI Architect
From mastering AI agents to building and deploying them at cloud scale — earn the Certified Agentic AI Architect credential across four levels.
5Quarters4Certification Levels11Exams
Includes Claude Certified Architect — Foundations (CCA-F) by Anthropic in Q4
Program Overview
The Learning Path
The Certified Agentic AI Architect is a five-quarter professional certification program for computer science practitioners and AI engineers. It takes you from AI-driven development fundamentals through building and deploying Agent Factories at cloud scale.
| Quarter | Title | Key Domains | Level Earned |
|---|---|---|---|
| AI-101 | AI-Driven Development, Claude Code, and OpenClaw Fundamentals | Claude CodeOpenClawAI AgentsMarkdown Specs | ☆ Contributes to Level 1 |
| AI-151 | Context Engineering, Spec-Driven Development, and Advanced Claude Code | Context EngineeringAgentic WorkflowsVersion Control | ★ Earns Level 1: Certified Agentic AI Foundations |
| AI-251 | Learn to Think and Program in the AI Era | PythonType-Driven AITestingOOPCLI & APIs | ★ Earns Level 2: Certified Agentic AI Developer |
| AI-321 | Building AI Agent Factory+ CCA-F (Anthropic) | Claude Agent SDKMCPFastAPIRAG+ CCA-F Domains | ★ Earns Level 3 + CCA-F |
| AI-451 | Deploying Agent Factories in the Cloud | DockerKubernetesKafkaCI/CDMulti-Cloud | ★ Earns Level 4: Certified Agentic AI Architect |
AI-101☆ Contributes to Level 1
AI-Driven Development, Claude Code, and OpenClaw Fundamentals
Claude CodeOpenClawAI AgentsMarkdown Specs
AI-151★ Earns Level 1: Certified Agentic AI Foundations
Context Engineering, Spec-Driven Development, and Advanced Claude Code
Context EngineeringAgentic WorkflowsVersion Control
AI-251★ Earns Level 2: Certified Agentic AI Developer
Learn to Think and Program in the AI Era
PythonType-Driven AITestingOOPCLI & APIs
AI-321★ Earns Level 3 + CCA-F
Building AI Agent Factory+ CCA-F (Anthropic)
Claude Agent SDKMCPFastAPIRAG+ CCA-F Domains
AI-451★ Earns Level 4: Certified Agentic AI Architect
Deploying Agent Factories in the Cloud
DockerKubernetesKafkaCI/CDMulti-Cloud
Anthropic Certification
Claude Certified Architect — Foundations (CCA-F)
Anthropic launched the CCA-F as its first official technical certification, backed by a $100 million investment in the Claude Partner Network. It is a proctored, scenario-based, architecture-level exam that validates whether a practitioner can design and ship production-grade Claude applications at enterprise scale.
In this program, the CCA-F serves as a supplementary certification in Quarter 4 (AI-321). By the end of Q4, students have covered all five CCA-F exam domains through the PIAIC curriculum. The CCA-F adds an external, industry-recognised credential from Anthropic that validates their production architecture skills alongside the internal PIAIC exams.
720 out of 1,000Passing Score
60Scenario-Based Questions
ProctoredExam Format
Each quarter earns you a progressive certification — see the ladder below.
Certification
Certification Ladder
Four progressive certification levels. Each earned by passing the corresponding quarter's exams. All four culminate in the Certified Agentic AI Architect credential.
1
Certified Agentic AI Foundations
2
Certified Agentic AI Developer
3
Certified Agent Factory Builder+ CCA-F (Anthropic)
4
Certified Agentic AI Architect
★
All Levels Completed → Certified Agentic AI Architect10 internal exams + CCA-F across 5 quarters
Explore each quarter's full syllabus below.
Curriculum
Curriculum Details
Five quarters of hands-on learning. Expand each quarter to see the full syllabus, assessments, and certification pathway.
AI-101
AI-Driven Development, Claude Code, and OpenClaw Fundamentals
☆ Contributes to Certified Agentic AI Foundations
Claude CodeOpenClawAI AgentsMarkdown SpecsCritical Thinking
Your entry point into the Agent Factory paradigm. Learn why AI is non-negotiable, master Claude Code as your primary development tool, build your first OpenClaw applications, and begin developing critical thinking skills that separate you from AI-dependent practitioners.
What You Will Learn
- The Agent Factory Thesis — the paradigm shift from writing software to manufacturing AI employees
- Which AI Employees Should You Use in 2026 — navigating the agent landscape
- Why AI Is Non-Negotiable — the economic imperative for every organisation
- The Agent Factory Paradigm — Digital FTEs, the Two-Layer Model, and the 10-80-10 Rule
- Markdown Writing Instructions — the foundational skill for spec-driven development
- General Agents — understanding agent architecture and capabilities
- OpenClaw Fundamentals — meet your first AI employee and build your first apps
- Asking Better Questions — question formulation that AI cannot replicate (Part 0)
- Detecting Broken Reasoning — spotting errors in AI-generated outputs (Part 0)
Assessments
Mid-Term:Agent Factory Intro Exam(L1: P1-AFI)
Final:Claude Code Foundation Exam(L1: P2-CCF)
Parallel Tracks
- ▸ OpenClaw Fundamentals — Online Classes and Hackathon
- ▸ Learn to Think Part 1 — Online Classes and Projects (Chapters 1–2)
Book Coverage: Preface + About + Thesis + Part 1: General Agents Foundations (Chapters 12–14) + Part 5: Building OpenClaw Apps + Part 0: Chapters 1–2
☆ Contributes to: Certified Agentic AI Foundations (contributes 2 of 4 exams)
AI-151
Context Engineering, Spec-Driven Development, and Advanced Claude Code
★ Earns Certified Agentic AI Foundations
Context EngineeringDatabasesAgentic WorkflowsSpec-Driven Dev
Master the art of context engineering and spec-driven development — the methodology that makes AI agents reliable. Build agent workflow primitives for file processing, structured data, and version control. Complete Level 1 certification.
What You Will Learn
- Context Engineering — the discipline of providing AI with exactly the right information
- Spec-Driven Development — writing specifications that agents can execute reliably
- The Seven Principles — foundational laws for building production-grade agent systems
- Claude Code for Teams & CI/CD — collaborative development and continuous integration
- File Processing Workflows — building agents that handle document pipelines
- Structured Data & Persistent Storage — agents that work with databases and state
- Version Control Workflows — agents that manage code repositories intelligently
- Thinking in Systems — seeing second-order consequences (Part 0)
- Reasoning From First Principles — knowing what to automate and what not to (Part 0)
- Communicating What Matters — stakeholder communication under pressure (Part 0)
Assessments
Mid-Term:Context Engineering, Spec-Driven Development, and Advanced Claude Code Exam(L1: P3-ACC)
Final:Agent Workflow Primitives Exam(L1: P4-AWP)
Parallel Tracks
- ▸ OpenClaw Project — Online Classes and Hackathon (Chapters 58–60)
- ▸ Learn to Think Part 2 — Online Classes and Projects (Chapters 3–5)
Book Coverage: Part 1: General Agents Foundations (Chapters 15–18) + Part 2: Agent Workflow Primitives + Part 5: OpenClaw Projects (Chapters 58–60) + Part 0: Chapters 3–5
★ Earns: Certified Agentic AI Foundations
AI-251
Learn to Think and Program in the AI Era
★ Earns Certified Agentic AI Developer
PythonType-Driven AITestingOOPCLI & APIs
The programming quarter. Eight phases take you from setting up the development workbench through type-driven development, testing, debugging, OOP, real-world Python, CLI applications, and production systems. In parallel, complete the remaining Thinking is the Curriculum chapters and assemble your Thinking Portfolio.
What You Will Learn
- Phase 1: The Workbench — setting up the professional development environment (Chapters 42–46)
- Phase 2: Specify with Types — types as the language of intent (Chapters 34–37)
- Phase 3: Tests as Specification — test-driven verification (Chapters 38–41)
- Phase 4: Debug & Master — debugging skills and TDG independence (Chapters 42–43)
- Phase 5: The Python Object Model — OOP done right (Chapters 44–47)
- Phase 6: Real-World Python — building practical applications (Chapters 48–50)
- Phase 7: CLI & Concurrency — command-line tools and parallel execution (Chapters 51–52)
- Phase 8: Production Systems — hardening and securing code (Chapters 53–54)
- Working With AI, Not For AI — the AI collaboration framework (Part 0)
- Reasoning Through Dilemmas, Building From Nothing, Deciding Under Uncertainty (Part 0)
- Learning How to Learn + Thinking Portfolio assembly (Part 0, Chapters 10–11)
Assessments
Mid-Term:Fundamentals of Programming in the AI Era(L2: P1-FPA)
Final:Advanced AI Programming in the AI Era(L2: P2-APA)
Parallel Tracks
- ▸ Learn to Think Part 3 — Online Classes and Projects (Chapters 6–11, completing Part 0)
Book Coverage: Part 4: Programming in the AI Era (Chapters 34–54, 8 Phases) + Part 0: Chapters 6–11
★ Earns: Certified Agentic AI Developer
AI-321
Building AI Agent Factory
★ Earns Certified Agent Factory Builder + CCA-F (Anthropic)
Claude Agent SDKOpenAI Agents SDKMCPFastAPIRAGCCA-F
The construction quarter. Build production-grade Agent Factories using the Claude Agent SDK and OpenAI Agents SDK. Master MCP server development, FastAPI integration, and RAG pipelines. At the end of this quarter, students also sit Anthropic’s official Claude Certified Architect — Foundations (CCA-F) exam as a supplementary industry certification.
What You Will Learn
- Claude Agent SDK — the primary framework for building custom agents
- OpenAI Agents SDK — comparative agent development framework
- Claude API — direct API integration for custom workflows
- Model Context Protocol (MCP) — building MCP servers and connectors
- FastAPI — production API endpoints for agent systems
- Retrieval-Augmented Generation (RAG) — grounding agents in organisational knowledge
- Agent Factory Architecture — the Factory Layer + Edge Layer pattern
- Multi-agent orchestration and handoff patterns
- CCA-F Exam Preparation — Anthropic Academy courses, scenario walkthroughs, and practice questions
Assessments
Mid-Term:Building Agent Factories Fundamentals(L3: P1-BAFF)
Final:Building Advanced Agent Factories(L3: P2-BAAF)
⭐ Supplementary:Claude Certified Architect — Foundations(CCA-F (Anthropic))
Book Coverage: Part 6: Building Agent Factories (complete) + Anthropic Academy CCA-F Preparation
⭐ Additional Certification
Claude Certified Architect — Foundations (CCA-F)
Anthropic's official proctored certification. 60 multiple-choice questions, proctored, scenario-based. Passing score: 720 out of 1,000. Students sit this exam at the end of Q4 as an external industry credential that supplements the internal PIAIC exams.
CCA-F Exam Domains & Curriculum Alignment:
Agentic Architecture & Orchestration (27%)
Covered in AI-321: Agent SDK, multi-agent orchestration, hub-and-spoke, lifecycle hooksTool Design & MCP Integration (18%)
Covered in AI-321 + AI-151: MCP servers, tool boundaries, resources, promptsClaude Code Configuration & Workflows (20%)
Covered in AI-101 + AI-151: CLAUDE.md, Agent Skills, plan mode, slash commands, CI/CDPrompt Engineering & Structured Output (20%)
Covered in AI-151 + AI-321: Context engineering, JSON schemas, extraction patterns, few-shotContext Management & Reliability (15%)
Covered in AI-151 + AI-321: Long-context, multi-agent handoffs, error propagation, escalationPreparation: 13 free courses on Anthropic Academy (anthropic.skilljar.com). Via Claude Partner Network (free to join).
6 Exam Scenarios:
Customer Support AgentCode Generation with Claude CodeMulti-Agent Research SystemDeveloper ProductivityClaude Code for CI/CDStructured Data Extraction
★ Earns: Certified Agent Factory Builder + Claude Certified Architect — Foundations (Anthropic)
AI-451
Deploying Agent Factories in the Cloud
★ Earns Certified Agentic AI Architect
DockerKubernetesKafkaCI/CDMulti-Cloud
The deployment quarter. Take your Agent Factories from development to production at cloud scale. Master containerisation, orchestration, event streaming, continuous deployment, and multi-cloud strategies.
What You Will Learn
- Docker — containerising agent systems for consistent deployment
- Kubernetes — orchestrating agent workloads at scale
- Kafka — event streaming for real-time agent communication
- CI/CD Pipelines — automated testing, building, and deployment
- Multi-Cloud Deployment — strategies for AWS, GCP, and Azure
- Production Monitoring — observability, logging, and alerting for agent systems
- Security & Compliance — securing agent deployments in enterprise environments
- Cost Optimisation — managing cloud spend for agent infrastructure
Assessments
Mid-Term:Agentic AI Cloud Deployment Fundamentals(L4: P1-AACDF)
Final:Advanced Agentic AI Cloud Deployment(L4: P2-AAACD)
Book Coverage: Part 7: Deploying Agent Factories in the Cloud (complete)
★ Earns: Certified Agentic AI Architect
What You'll Learn
Skills That Define the AI-Native Era
Master the skills to build, deploy, and monetize AI systems — from your first agent to your first AI business.
Agentic AI Systems
Build autonomous agents that reason, plan, and use tools. Orchestrate multi-agent workflows with memory, retrieval, and real-world integrations.
AI Employees & Automation
Build and deploy always-on AI employees that work for you while you sleep. Automate workflows and scale from personal assistant to enterprise-grade systems.
Python & TypeScript Mastery
Python for intelligence and backend power. TypeScript for interaction and UI. Master the bilingual stack that drives the AI-native ecosystem.
Cloud-Native Deployment
Containerize and deploy agent systems at scale. Master CI/CD pipelines, observability, and multi-cloud infrastructure for production-grade reliability.
AI Business & Monetization
Transform your AI skills into revenue streams. Build subscription services, license expertise as skill packs, or sell on agent marketplaces.
Voices on AI
Will AI Replace Humans — and Is Pakistan Ready?
Zia Khan sits down with Farhan Mallick on Raftar Now for a candid conversation about automation, the skills that still matter, and what it takes for Pakistan to stay competitive as work goes AI-native — themes that sit at the center of PIAIC's national upskilling mission.
Zia Khan
COO PIAIC · CEO Panacloud
"Our mission is to build Pakistan's future-ready workforce — equipped to create AI agents that think, act, and generate value autonomously."
Tamgha-e-ImtiazHonorary Ph.D.8x Microsoft MVP
Ready to Start?
The Future Belongs to
Agent Builders
Join 17,000+ learners across Pakistan. Free enrollment. World-class curriculum. Build AI agents that generate real revenue.