Shrutam Free CCA-F Prep · Crash Tier BETA

Anthropic Claude Certified Architect — Foundations

Free Hinglish-narrated prep for the official CCA-F exam. Built by senior Indian IT-services architects, for senior Indian IT-services architects.

Exam format
60 MCQs
Duration
120 min
Pass mark
720 / 1000
Domains
5 (weighted)
Cost
$99 (free for partners)
Beta progress: 50/50 Hinglish lessons live · 26/50 English lessons live · 2 mock exams live. Default language on this domain: Hinglish.

Domain 1 Agents & Workflows 27%

d1-l01-agent-definition
What is an Agent? — Anthropic's Definition
~12 min · HG ✓ EN ✓
d1-l02-hub-and-spoke
Hub-and-Spoke vs Sequential Orchestration
~12 min · HG ✓ EN ✓
d1-l03-react-loop
The ReAct Loop — Thought → Action → Observation
~12 min · HG ✓ EN ✓
d1-l04-bounded-react
Bounded Iteration — Why max_iterations Matters
~12 min · HG ✓ EN ✓
d1-l05-multi-agent-research
Multi-Agent Research Pipeline — Planner + Workers + Critic
~12 min · HG ✓ EN ✓
d1-l06-oauth-token-relay
Context Handoff Between Agents (OAuth-Relay Pattern)
~12 min · HG ✓ EN ✓
d1-l07-confidence-gate
Confidence Thresholds + Human-in-the-Loop Escalation
~12 min · HG ✓ EN ✓
d1-l08-task-decomposition
Task Decomposition — When to Split, When to Merge
~12 min · HG ✓ EN ✓
d1-l09-state-management
Agent State Management — Stateless API, Stateful Workflow
~12 min · HG ✓ EN ✓
d1-l10-parallel-tools
Parallel Tool Calls vs Sequential — The Latency Math
~12 min · HG ✓ EN ✓
d1-l11-router-pattern
Router Pattern — How a Lead Agent Decides
~12 min · HG ✓ EN ✓
d1-l12-dag-with-contracts
DAG Orchestration — Manhattan Subway Map
~12 min · HG ✓ EN ✓
d1-l13-recursive-delegation
Recursive Sub-Task Delegation
~12 min · HG ✓ EN ✓
d1-l14-safety-rails
Agentic Safety Rails — Anthropic's Guardrails Pattern
~12 min · HG ✓ EN ✓

Domain 2 Tools & Model Context Protocol 20%

d2-l01-mcp-overview
What is MCP? — Model Context Protocol in 12 Minutes
~12 min · HG ✓ EN ✓
d2-l02-mcp-architecture
MCP Host / Client / Server — The Three Layers
~12 min · HG ✓ EN ✓
d2-l03-read-resource-write-tool
Resource vs Tool — The Read/Write Distinction
~12 min · HG ✓ EN ✓
d2-l04-service-mesh-gateway
Tool Boundary Discipline — Why Tools Shouldn't Reason
~12 min · HG ✓ EN ✓
d2-l05-tool-output-size
Tool Output Size — Avoiding Context Blow-up
~12 min · HG ✓ EN ✓
d2-l06-stdio-vs-sse
MCP Transport — stdio vs SSE vs HTTP
~12 min · HG ✓ EN ✓
d2-l07-explicit-tool-choice
tool_choice — auto, explicit, and forced
~12 min · HG ✓ EN ✓
d2-l08-tool-error-handling
Error Propagation in Tool Responses
~12 min · HG ✓ EN ✓
d2-l09-mcp-production
Building a Production MCP Server — Auth, Logging, Versioning
~12 min · HG ✓ EN ✓

Domain 3 Claude Code in Production 20%

d3-l01-claude-code-overview
Claude Code 101 — CLI Mental Model
~12 min · HG ✓ EN ✓
d3-l02-enterprise-override
CLAUDE.md Hierarchy — Project / User / Enterprise Cascade
~12 min · HG ✓ EN ✓
d3-l03-settings-permissions
settings.json — Permissions, Hooks, MCP Servers
~12 min · HG ✓ EN ✓
d3-l04-pretooluse-hook
PreToolUse / PostToolUse Hooks — Deterministic Gates
~12 min · HG ✓ EN ⏳
d3-l05-isolated-subagent
Subagent Dispatch — Isolated Context Bubbles
~12 min · HG ✓ EN ⏳
d3-l06-slash-commands
Custom Slash Commands — Your Team's Vocabulary
~12 min · HG ✓ EN ⏳
d3-l07-plan-vs-edit
Plan Mode vs Edit Mode — When to Stay Read-Only
~12 min · HG ✓ EN ⏳
d3-l08-multi-file-refactor
Multi-File Migrations — Refactoring at Scale
~12 min · HG ✓ EN ⏳
d3-l09-claude-code-cicd
CI/CD with Claude Code — Headless Mode + Outputs
~12 min · HG ✓ EN ⏳
d3-l10-subagent-frontmatter
Subagent Definitions — Custom Roles via Frontmatter
~12 min · HG ✓ EN ⏳

Domain 4 Prompt Engineering for Architects 18%

d4-l01-prompt-engineering-overview
Anthropic's Prompt Engineering Philosophy
~12 min · HG ✓ EN ⏳
d4-l02-xml-scoped-sections
XML Tag Scoping — Sections That Don't Bleed
~12 min · HG ✓ EN ⏳
d4-l03-system-prompt-design
System Prompts — Role, Rules, Constraints
~12 min · HG ✓ EN ⏳
d4-l04-examples-before-task
Few-Shot Examples — Placement and Count
~12 min · HG ✓ EN ⏳
d4-l05-structured-output-mode
Structured Output Mode — JSON Schema Enforcement
~12 min · HG ✓ EN ⏳
d4-l06-extended-thinking-output-tag
Extended Thinking — When to Enable, When to Skip
~12 min · HG ✓ EN ⏳
d4-l07-validation-retry
Retry-on-Validation-Fail Loops
~12 min · HG ✓ EN ⏳
d4-l08-negative-constraints
Negative Constraints — Why 'Don't' Fails
~12 min · HG ✓ EN ⏳
d4-l09-persona-prompts
Persona Prompts — When They Help and When They Hurt
~12 min · HG ✓ EN ⏳
d4-l10-pydantic-integration
Pydantic Integration — End-to-End Typed Pipelines
~12 min · HG ✓ EN ⏳

Domain 5 Production Operations & Cost 15%

d5-l01-context-window-economics
The Context Window — 200K Tokens Isn't Free
~12 min · HG ✓ EN ⏳
d5-l02-cache-static-prefix
Prompt Caching — Where to Place Cache Breakpoints
~12 min · HG ✓ EN ⏳
d5-l03-chapter-boundary-compaction
Long-Context Compaction — Chapter Boundaries vs Token Cuts
~12 min · HG ✓ EN ⏳
d5-l04-exponential-jitter
Rate Limits and Exponential Backoff
~12 min · HG ✓ EN ⏳
d5-l05-right-size-model
Model Routing — Haiku for Speed, Sonnet for Reasoning, Opus for Architecture
~12 min · HG ✓ EN ⏳
d5-l06-telemetry-logging
Telemetry and Observability — What to Log
~12 min · HG ✓ EN ⏳
d5-l07-reliability-patterns
Reliability Patterns — Retries, Fallbacks, Circuit Breakers
~12 min · HG ✓ EN ⏳

Mock Exams

60 scenario-anchored questions each, balanced across all 5 domains by official exam weighting. Founder-vetted question bank. Scored 100-1000, pass mark 720.

Mock Exam 1

60 questions · ~120 min · Authentic Indian enterprise scenarios

Start Mock 1

Mock Exam 2

60 questions · ~120 min · Different bank, same difficulty mix

Start Mock 2