Library › Software developers

AI exercises for software developers

50 exercises built around real tasks. Each takes about 5 minutes.

• Add dark mode by overriding your existing CSS variablesCoding · Setting limits and rulesPremium
• Add input validation to a Java method with exact messagesCoding · Giving the right contextPremium
• Add type hints that still run on Python 3.8Coding · Giving the right contextPremium
• Check an AI's API answer against the real docsEvaluating AI · Catching mistakes and made-up factsPremium
• Convert a Java loop to a stream without losing key orderCoding · Setting limits and rulesPremium
• Convert OFFSET paging to keyset paginationCoding · Setting limits and rulesPremium
• Count returns per category without double-counted rowsCoding · Giving the right contextPremium
• Decide if an LLM judge is good enough to gate CIEvaluating AI · Catching mistakes and made-up factsPremium
• Explain a background job to your PM without jargonCoding · Saying who it's forPremium
• Explain a SQL query to an analyst in 4 labeled linesCoding · Saying who it's forPremium
• Extract a pure function so the fee logic can be testedCoding · Setting limits and rulesPremium
• File a bug report when an AI PR summary misses a breaking changeEvaluating AI · Asking for a format you can usePremium
• Fix a KeyError from a stack trace without touching anything elseCoding · Setting limits and rulesPremium
• Fix a shell loop that breaks on filenames with spacesCoding · Setting limits and rulesPremium
• Fix an async forEach without breaking the rate limitCoding · Giving the right contextPremium
• Fix one line that set -e kills, and keep strict modeCoding · Setting limits and rulesPremium
• Get each ticket's latest status on MySQL 8 without DISTINCT ONCoding · Giving the right contextPremium
• Grade 4 AI commit messages against your team's conventionEvaluating AI · Asking for a format you can usePremium
• Group 13 failing code-assistant runs into 3 causesEvaluating AI · Asking for a format you can usePremium
• Pick the metric that shows if AI-written tests catch bugsEvaluating AI · Giving the right contextPremium
• Refactor an if-else chain into a lookup map, plus one new ruleCoding · Setting limits and rulesPremium
• Refactor nested null checks into an Optional chainCoding · Setting limits and rulesPremium
• Replace a racy read-then-write with a single upsertCoding · Giving the right contextPremium
• Review a migration with one MUST, one SHOULD and one NITCoding · Asking for a format you can usePremium
• Review an Express handler as 4 labeled line commentsCoding · Asking for a format you can usePremium
• Turn "the review bot is noisy" into 4 testable checksEvaluating AI · Catching mistakes and made-up factsPremium
• Turn a rate-limit spec into a typed TypeScript stubCoding · Setting limits and rulesPremium
• Update equals and hashCode together for a new ruleCoding · Giving the right contextPremium
• Write 3 named JUnit 5 tests with plain assertionsCoding · Setting limits and rulesPremium
• Write 5 edge-case tests for an internal docs Slack botEvaluating AI · Giving the right contextPremium
• Write a Bash usage message that matches the real defaultsCoding · Keeping the AI to your sourcesPremium
• Write a bug report for a summary that dropped a chargeEvaluating AI · Asking for a format you can usePremium
• Write a JSON judge for replies in the wrong languageEvaluating AI · Setting limits and rulesPremium
• Write a JSON judge that catches made-up SQL columnsEvaluating AI · Setting limits and rulesPremium
• Write a JSON judge that catches personal data in summariesEvaluating AI · Setting limits and rulesPremium
• Write a LEFT JOIN that keeps customers who spent nothingCoding · Giving the right contextPremium
• Write a PR description in your team's templateCoding · Asking for a format you can usePremium
• Write a Python log regex with exact group namesCoding · Setting limits and rulesPremium
• Write a README config table from the code, not the old docsCoding · Keeping the AI to your sourcesPremium
• Write a regression checklist before changing a review botEvaluating AI · Giving the right contextPremium
• Write a regression checklist before shrinking RAG chunksEvaluating AI · Giving the right contextPremium
• Write a revert commit message in your team's formatCoding · Asking for a format you can usePremium
• Write a safe duplicate-cleanup DELETE you can dry-runCoding · Giving the right contextPremium
• Write a SKU regex that passes the tests plus a new ruleCoding · Giving the right contextPremium
• Write a TypeScript type guard without a validation libraryCoding · Setting limits and rulesPremium
• Write an argparse parser from a CLI spec, and nothing moreCoding · Setting limits and rulesPremium
• Write exactly 3 named Jest tests for a slugify helperCoding · Setting limits and rulesPremium
• Write one crontab line that won't overlap or lose logsCoding · Giving the right contextPremium
• Write one production-safe index from an EXPLAIN planCoding · Giving the right contextPremium
• Write the golden answer for a dev-helper bot's git questionEvaluating AI · Giving the right contextPremium