Course Overview
Most AI coding courses stop once participants can prompt a tool into building a prototype. This course goes further: participants build a small resident coding and workflow agent in Claude Code or Codex that can remember project context, use tools, run on a schedule, keep logs, and pause for human approval when the work gets risky.
The course is designed for participants who have already seen the baseline ideas behind AI coding: prompt and context engineering, repo instructions such as AGENTS.md or CLAUDE.md, file search, tool choice, skills, MCPs, subagents, code review, testing, documentation, and worktrees. We revisit those ideas through one coherent build instead of treating them as disconnected revision slides.
We also touch on the emerging shape of agentic engineering: designing loops that find work, hand it to agents, capture state outside the chat, separate maker and checker roles, and bring a human back in when judgement matters.
The build can be framed around a theme such as Humanitarian Assistance and Disaster Relief (HADR), where an agent researches a developing incident, monitors selected public sources, prepares cited updates, and generates reports. The exact theme may vary, but the pattern stays the same: a real workflow with sources, changing context, outputs, logs, tests, diagrams, and approval gates that a human can inspect.
Lesson Outcomes
By the end of this course, participants will have built and reviewed a working mini-agent system that can:
Maintain useful project memory. Store repo instructions, task notes, decisions, source summaries, and reusable operating procedures without stuffing every prompt with stale context.
Use tools with clear boundaries. Decide what belongs in a skill, what belongs in an MCP or tool integration, and what should stay as plain repo instructions.
Run a research and monitoring workflow. Collect source material, track changes, produce situation updates, and generate a final report with citations and human review points, using HADR or another comparable theme as the course context.
Control token and model use. Route planning, execution, review, and documentation tasks to suitable models; reset context deliberately; and keep skills and instructions lean.
Add production guardrails. Apply sandboxing, credential handling, prompt-injection checks, approval gates, audit trails, and rollback plans to an agentic workflow.
Document the system for human inspection. Generate tests, review notes, runbooks, and C4-style system views that make the agent's behaviour easier to check.
Curriculum
Day 1: Planning and familiarisation
Participants get comfortable directing an agent through short builds before planning the main course project: a monitoring agent that reads live sources, assesses new events, and publishes a useful dashboard.
Set up Claude Code or Codex, the working repository, project instructions, and a GitHub-based review workflow.
Build a small API app and a CRUD app to practise planning, implementation, testing, and review with an agent.
Use repo instructions such as
CLAUDE.mdorAGENTS.md, plan mode, worktrees, and pull requests to keep agent work bounded and inspectable.Define the monitoring agent's users, sources, outputs, constraints, and vertical slices in a practical product requirements document.
Day 2: Harnesses, loops, and agents
Participants turn the project from a prompt-driven app into a resident agent. They examine the anatomy of an agentic system—standing instructions, loop, tools, file-based memory, heartbeat, and output channel—then build those parts progressively.
Challenge the product plan, file the first vertical slices as issues, and build an end-to-end slice in a worktree with pull-request review.
Build a compact Python agent harness, progressing from a chat loop and standing instructions to tool calls, an autonomous loop, and dashboard output.
Compare the same task across a hand-built harness, Claude Code or Codex, and another coding harness to separate the role of the model from the surrounding system.
Evaluate when to use skills, MCP integrations, command-line tools, APIs, headless runs, and different models.
Day 3: Make it trustworthy, then ship it
The final day adds the controls needed to let the agent work with less supervision. Participants finish the automation, run a bounded task in the background, publish the product, and improve it through structured peer feedback.
Add deterministic hooks, reviewer subagents, checkpoints, persistent memory, and context and token hygiene.
Turn repeated instructions into reusable skills and add a heartbeat using GitHub Actions, a server, or a local scheduler.
Launch a sandboxed background run with explicit success checks and hard limits on time, permissions, and spend.
Publish the dashboard and its supporting site, collect feedback as prioritised issues, work the backlog, and present the finished system.
Details
Dates and Times: 7-9 Jul 2026 (3 full-day sessions), or 12-13 Aug 2026 (2 full-day sessions).
Location: CT Hub 2, Lavender.
Requirements: Bring your own laptop. Modern Mac, Windows, or Linux laptops are supported. Participants should be comfortable using a terminal, Git, GitHub, and at least one AI coding tool such as Claude Code or Codex.
Prerequisites: Participants should have completed AIxTech or have equivalent hands-on experience with AI coding tools, prompt engineering, context engineering, basic software review, and testing/debugging.
Tools and accounts: The course uses Claude Code or Codex as the main coding harness, with comparisons to other agentic coding tools where useful. The course setup provides six months of access to the selected 5x plan, using either Claude Code or Codex depending on final setup.
Fees (excl. GST): S$1,760 per participant for the 3-day course (limited-time pricing, regular $2010), or S$1,710 per participant for the condensed 2-day course. Minimum 10 participants to start.
All quoted prices are in SGD and are before 9% GST. Invoicing terms available.