snap
Sign in
ANDMOD.AND-10 · v1.0

Ship code 5× faster
— without 5×
the bugs.

10 micro-lessons · ~102 min · Real Docker images

THE SHIPPING BAY · LIVE
BAY.A · 4 ROUTES
ROUTING
REPOPLANEDITTESTSHIP
ROUTE 1/4 · in-flightagent + tests
ANDROLE TRACKHOT

AI-native software development

Ship 5× faster with coding agents you actually trust.

WHY THIS MATTERS · SWE-BENCH VERIFIED · ANTHROPIC 2026 AGENTIC CODING TRENDS · VECTARA 591-INCIDENT TAXONOMY
By April 2026, top SWE-bench Verified scores cleared 87.6% (Claude Opus 4.7 Adaptive). Stripe ships 1,000+ AI PRs/week internally; Shopify reports 20% engineering productivity lift. The pattern that makes those numbers real isn't 'better prompts' — it's the read-plan-edit-verify loop with rules, hooks, MCPs, and a cost dashboard. This course is the rollout playbook.
WHAT YOU'LL LEARN
01The agentic dev loop
02Project rules: AGENTS.md / CLAUDE.md
03Plan mode and the read-only stage
04TDD with agents — the only way that scales
05Subagents and parallel research
06MCP servers and tool wiring
07Hooks, automations, and rails
08PR review automation and auto-fix loops
09Cost, telemetry, and $-per-merged-PR
10Production rollout: SOC 2, secrets, and audit
YOU'LL BE ABLE TO
Run the read-plan-edit-verify loop reliably across claude-code / cursor-agent / aider
Author CLAUDE.md / AGENTS.md, hooks, and MCP servers that survive long sessions
Wire a TDD-with-agent loop, PR review pipeline, and $/merged-PR dashboard end-to-end
Run a Tier 0 → Tier 1 → Tier 2 rollout with audit log + signing + kill switch
Cut $-per-merged-PR by 5-10× via structural fixes (rules, hooks, severity tagging)
SKILLS YOU'LL GAIN

Real skills, real career delta.

Skills you'll gain

10
  • Diagnose when to leave autocomplete for an agent loopWorking

    Use the 3-signal test (ticket length, cross-file edits, test surface) to pick autocomplete vs agent before opening the IDE — and quote the cost difference.

  • Author project rules (CLAUDE.md / AGENTS.md / cursor.rules)Production

    Write rules that survive 8-hour sessions: build commands, conventions, DO/DO-NOT lists, gotchas, hook config — distilled from your bug history, not copied from a template.

  • Run plan-then-edit reliably across agent CLIsWorking

    Use plan mode in Claude Code, plan-as-edit in Cursor, plan-files in Aider, PRD generation in Devin — recognise when to skip plan and when it's mandatory.

  • Drive a TDD-with-agent loop that convergesProduction

    Red-green-refactor in agent loop: failing test as input, agent edits until green, agent stops at green. Includes the 'one bug = one regression test' commit rule.

  • Dispatch parallel sub-agents for read-only researchProduction

    Spawn N independent Explore agents on non-overlapping queries; aggregate results back to a primary agent. Cost math: 4× tokens parallel often beats 4× sequential time at the team level.

  • Wire MCP servers (Linear, Sentry, Postgres, custom) into agentsWorking

    Install standard MCP servers, write a custom one in <30 lines with FastMCP, scope read-only vs read-write, pair it with a permission boundary.

  • Configure agent hooks for lint / type / secrets enforcementProduction

    Pre-tool-use, post-edit, post-stop hooks that catch unwanted edits BEFORE they ship. Hooks beat prompts because they don't depend on the model remembering.

  • Ship PR-review automation with auto-fix loopsProduction

    CodeRabbit / Greptile / self-hosted reviewer + agent that consumes the review, fixes, re-runs tests. The 'no human-in-loop until tests pass' pattern.

  • Track $-per-merged-PR and route work by itProduction

    Helicone / OpenAI dashboard / Anthropic Console + Slack digest of weekly $/PR by team. Use it to settle 'which agent / which model / which mode' debates with data.

  • Operate AI-native dev under SOC 2 / GDPR / regulated constraintsAdvanced

    Permission models, audit logs, secrets scanning preflight, sandboxed exec, allow-listed shell — the rollout shape regulated industries actually buy.

RUNNABLE ON YOUR MACHINE
$ docker pull snap/ai-native-dev:agent-dev-shell
$ docker run --rm -it snap/ai-native-dev:agent-dev-shell
snap/ai-native-dev:agent-dev-shell
QUICK PREVIEW · 7 MIN
VERIFIED ENGINEER REVIEWS
The agent-dev-shell bake-off is the spike kit my team has been missing. We picked claude-code for backend and cursor for frontend in week 2.
@claude_code_kimVERIFY ON GITHUB
The cost-dashboard lesson alone paid for the course. We dropped from $4.20 median $/PR to $0.30 in 8 weeks of running it.
@dx_eng_julesVERIFY ON GITHUB
LESSONS10
HOURS~1.7
LEARNERS0
THIS WEEK+0%