Course Outline
From Autocomplete to Agents: Understanding Failure Modes
• Structure of a coding agent: model, harness, tool surface, context, and permissions
• Positioning of various tools: Claude Code, GitHub Copilot, Cursor, Codex CLI, and Gemini CLI
• Categorization of failure types: incorrect context, unsuitable tools, lack of feedback, and unbounded autonomy
Demonstration: Executing the same task under optimal and suboptimal conditions, compared side-by-side
Context Engineering
• Treating the context window as a limited budget: determining what deserves inclusion
• AGENTS.md, CLAUDE.md, .cursor/rules, copilot-instructions.md — distinct filenames serving a single source of truth
• Defining conventions, build and test commands, and architectural boundaries
• Retrieval vs. explicit context; task decomposition and sub-agents
Lab: Create repository context for an unfamiliar Python service, then re-run a previously failing task to compare results
Reusable Workflows and Agent Skills
• Selecting the right abstraction: instruction files, skills, custom commands, or plain scripts
• Structure of a skill: triggering mechanisms, instructions, bundled scripts, and progressive disclosure
• Ensuring portability across tools and identifying points of potential lock-in
• Versioning, review, and team-wide distribution; avoiding common anti-patterns
Lab: Develop and test a reusable workflow that enforces a specific house coding standard
MCP: Linking Agents to Real-World Systems
• Architecture overview: clients, servers, tools, resources, and prompts; stdio and HTTP transports
• Identifying high-value servers: Git hosting, issue trackers, databases, browsers, and internal APIs
• Scenarios where a CLI or script outperforms an MCP server
• Maintaining tool-surface hygiene: understanding why excessive tools reduce reliability
Lab: Configure MCP servers to process a ticket end-to-end — issue, branch, patch, tests, and pull request
Feedback Loops and Evaluation
• Using tests, types, and linters as ground truth; employing test-first work as a control mechanism
• Leveraging CI as the outer loop, and applying review discipline to agent-generated diffs
• Creating golden-task evaluation sets: defining metrics and detecting regressions
• Treating cost and latency as primary performance indicators
Lab: Construct a small evaluation set and score two different agent configurations against it
Security and Guardrails
• Preventing prompt injection via issues, pull requests, READMEs, dependencies, and fetched pages
• Implementing permission models: allowlists, approvals, read-only tools, and network egress control
• Managing secret hygiene and sandboxing: using containers, ephemeral credentials, and limiting blast radius
• Assessing supply-chain risks in third-party MCP servers and shared skills
Lab: Observe an agent compromised by a poisoned repository, then harden the setup to prevent such incidents
Team-Level Implementation
• Defining a staged adoption path; deciding what to standardize and what to leave to individual preference
• Identifying metrics that reflect true value versus those that do not
Requirements
• Proficiency in Python, Git, and command-line interfaces
• Prior experience with an AI coding assistant
• NobleProg will provision Dadesktop VMs for participants, pre-installed with Docker, VS Code, and Python 3.11 or later
• A functional AI coding assistant of the participant's preference: Claude Code, GitHub Copilot, Cursor, Codex CLI, or Gemini CLI. The labs are tool-agnostic, with instructions provided for each platform
Target Audience
• Software engineers, tech leads, and architects seeking reliable outcomes from AI coding assistants
• Platform and developer-experience engineers implementing AI tooling across teams
• Engineering managers defining standards, guardrails, and success metrics
Testimonials (3)
Strong confirmation that current LLMs are deployable for customer projects when the proper setup is in place.
Oliver Fromm
Course - Agentic AI for Enterprise Applications
Machine Translated
The trainer is patient and very helpful. He knows the topic well.
CLIFFORD TABARES - Universal Leaf Philippines, Inc.
Course - Agentic AI for Business Automation: Use Cases & Integration
The mix of theory and practice and of high level and low level perspectives
