Applied AIwith Sawan Ruparel

Schedule and materials

The 15-week course sequence, lecture notes, slide decks, and suggested practice.

Public self-study edition

Follow the weeks in order or use them as modules at your own pace. Each lecture has a reading view, a canonical production failure mode to inspect, and its original slide deck.

In the taught university offerings, the course was assessed through three graded phase assignments (at the end of Phases 1, 2, and 3) followed by the capstone project (spanning Phases 4 and 5). In this public edition, use those phase assignments and decision artifacts as self-study checkpoints feeding into the Week 15 capstone. Check MCP version compatibility before the server exercises.

Foundations and reasoning

Understand model capabilities and inference-time scaling, engineer reliable structured prompts, design deterministic state machines before adding autonomy, and represent domain knowledge with embeddings and hybrid indexes.

WeekTopic and suggested practiceMaterials
Week 1

The AI model landscape

Reasoning models, inference-time compute, model selection, and the limits of benchmarks.

Production failure mode: Confident reasoning over fabricated premises

Suggested practice

Compare two available models on the same task. Record model versions, outputs, latency, and the limits of your comparison.

Week 2

Prompt engineering and structured outputs

Prompt design, structured responses, tool use, prompt versioning, and reliability.

Production failure mode: Unconstrained chain-of-thought overthinking and post-hoc rationalization

Suggested practice

Create a versioned prompt with a structured output schema and an evaluation set. Compare it with a simple baseline.

Week 3

The AI developer stack

Framework choices, agent loops, observability, error handling, and deployment patterns.

Production failure mode: Premature agent-ification and opaque framework wrappers

Suggested practice

Build a tool-using agent with two tools, structured output, and at least 10 scored evaluation cases. The lecture uses PydanticAI.

Week 4

Knowledge representation

Embeddings, vector search, hybrid retrieval, knowledge graphs, and context design.

Production failure mode: Bi-encoder top-k similarity without metadata filtering or reranking

Suggested practice

Index a small corpus. Compare keyword, vector, and hybrid retrieval on the same hand-labeled questions. Compile the Phase 1 Baseline, Control-Flow & Indexing Spec.

Knowledge augmentation and evaluation

Architect hybrid and graph-based retrieval pipelines, build automated evaluation harnesses, and measure faithfulness before scaling complexity.

WeekTopic and suggested practiceMaterials
Week 5

Retrieval-augmented generation

Chunking, retrieval, reranking, grounding, and iterative retrieval workflows.

Production failure mode: Conflating retrieval misses with generation hallucinations

Suggested practice

Build a RAG baseline with citations. Track retrieval failures separately from answer-generation failures.

Week 6

GraphRAG and global context

Graph-based retrieval, community summaries, global questions, and evaluation tradeoffs.

Production failure mode: Deploying GraphRAG on low-degree corpora or point-lookup workloads

Suggested practice

Compare graph-based retrieval with the Week 5 baseline on both local and corpus-wide questions.

Week 7

Evaluation and LLM-as-a-judge

Evaluation layers, RAG metrics, judge biases, calibration, and human review.

Production failure mode: Uncalibrated LLM-as-a-Judge position, verbosity, and self-preference bias

Suggested practice

Evaluate two retrieval or prompt variants on 50 questions. Report faithfulness, relevance, context precision, and context recall. Produce the Phase 2 Retrieval & Eval Harness Report.

MCP and agentic systems

Standardize tool integration with the Model Context Protocol (MCP), coordinate multi-step agentic workflows, engineer failure recovery, and design human-in-the-loop escalation.

WeekTopic and suggested practiceMaterials
Week 8

Model Context Protocol

Tools, resources, prompts, transport, authorization, and host/server responsibilities.

Production failure mode: Context bloat from inline tool dumps and unconstrained schemas

Suggested practice

Inspect a tool call and document its schema, authorization requirements, and failure behavior. Identify the protocol version used.

Week 9

Building MCP servers

Server implementation, validated tool handlers, testing, remote deployment, and operational safety.

Production failure mode: Swallowing business errors as transport exceptions (or vice versa)

Suggested practice

Implement a small MCP server against a disposable data source. Test invalid inputs, rejected access, and real tool calls.

Week 10

Multi-agent workflows

Coordination patterns, shared state, tool isolation, cost controls, and failure recovery.

Production failure mode: Unbounded retry loops and cascading failure across cooperating agents

Suggested practice

Compare a single-agent baseline with one coordinated workflow. Measure task completion, latency, and cost. Draft the capstone proposal.

Week 11

Human oversight and memory

Approval gates, escalation, persistent state, memory retrieval, and privacy.

Production failure mode: Uncheckpointed autonomous side effects and stale memory poisoning

Suggested practice

Add an approval gate and durable state to a workflow. Test rejected actions, restart behavior, and deletion of stored information. Complete the Phase 3 MCP Tool Contract & Recovery Trace.

Efficiency and governance

Optimize unit economics and latency via small language models (SLMs) and routing, distill or fine-tune deliberately against a baseline, and red-team adversarial failure modes.

WeekTopic and suggested practiceMaterials
Week 12

Small models and model routing

Task-specific benchmarking, quantization, serving, hardware limits, and model cascades.

Production failure mode: Miscalibrated router confidence sending edge cases to the fast tier

Suggested practice

Benchmark three model sizes on a 200-example task set and plot cost against quality. Collect the capstone's first evaluation results.

Week 13

Distillation and fine-tuning

Synthetic data, LoRA, QLoRA, preference tuning, evaluation, and data provenance.

Production failure mode: Teacher-error amplification and near-duplicate leakage in synthetic SFT data

Suggested practice

Design a small fine-tuning experiment with a held-out test set and documented data permissions. Compare with the untuned baseline.

Week 14

Safety and red-teaming

Prompt injection, agent hijacking, guardrails, least privilege, and adversarial evaluation.

Production failure mode: Indirect prompt injection hijacking privileged agent tools

Suggested practice

Add layered controls to a chatbot or agent. Test both blocked attacks and legitimate tasks; record failures, complete the Phase 4 Pareto Frontier & Red-Team Audit, and freeze the capstone demo.

Capstone

Build a reproducible, end-to-end AI system and defend its architectural decisions with quantitative evaluation evidence.

WeekTopic and suggested practiceMaterials
Week 15

Capstone: build, measure, explain

A working AI system, architecture writeup, measured evaluation, and an honest failure analysis.

Production failure mode: Reporting aggregate scores without a categorized failure taxonomy

Suggested practice

Deliver reproducible code, a 15-minute demonstration, a 4–6 page writeup, and evaluation results. Use the published capstone rubric for self-assessment.