spec
The specification of qualifications and characteristics an employer requires for a job role.
As defined in Roland Fryer on Race, Diversity, and Affirmative Action
Other definitions
A detailed written specification of what an application should be and do, including technical details, database schema, and API endpoints, used to instruct the coding agent.
Vibe Coding Tutorial and Best Practices (Cursor / Windsurf)
Where it comes up · 4
- Stop Prompting Claude. Use Karpathy's Method Instead.12 claims
- Roland Fryer on Race, Diversity, and Affirmative Action2 claims
- Vibe Coding Tutorial and Best Practices (Cursor / Windsurf)2 claims
- Lex Fridman Podcast #501 - DHH1 claim
Most important claims · 10 of 17Search all claims for “spec” →
- Using cursor/windsurf rules files (project-level natural-language instructions) to dictate how the agent codes, which technologies to use, and which workflows to follow is critical—especially as the codebase grows—because the agent otherwise introduces technologies not in your spec and breaks other things when fixing bugs.Vibe Coding Tutorial and Best Practices (Cursor / Windsurf)
- Writing a very detailed spec first (a thorough explanation of what you want built, including technical specification, database schema, and API endpoints) before pasting it into the coding agent produces better results.Vibe Coding Tutorial and Best Practices (Cursor / Windsurf)
- A large financial institution insisted it couldn't be biased because it only considered candidates with a 3.8 GPA in Applied Math from an Ivy League school, yet none of its own executives had that background—demonstrating that stated specs often bear no relation to the paths that actually produced the firm's leaders.Roland Fryer on Race, Diversity, and Affirmative Action
- Across a couple dozen employers, none had the right job spec: when you correlate collected applicant characteristics with actual performance and compare the most predictive ones to what the spec actually asks for, they do not line up—so disparities can arise from supply and mis-specified requirements rather than biased screening.Roland Fryer on Race, Diversity, and Affirmative Action
- A spec is a detailed document that translates your understanding and contextual information into a format Claude can use; Karpathy advocates moving beyond high-level plan mode to work with AI agents to design much more detailed, compartmentalized specs.Stop Prompting Claude. Use Karpathy's Method Instead.
- Layer 1 (spec) and Layer 2 (verifier) require somewhere to live, which is Layer 3 (the environment); the environment is a persistent workspace system (like a workshop with a blueprint pinned to the wall, a quality check station by the door) that allows the system to function at a high level.Stop Prompting Claude. Use Karpathy's Method Instead.
- The goal of any task is the conclusion you are trying to draw or the decision the task drives, which is fundamentally something AI will never be able to decide on its own; uncovering the actual goal is the first step to creating an effective spec.Stop Prompting Claude. Use Karpathy's Method Instead.
- The more precise you are in your spec instructions, the less AI has to assume; every assumption AI makes is a chance for it to drift from the final product you want, so precision directly correlates with output fidelity.Stop Prompting Claude. Use Karpathy's Method Instead.
- The complete Karpathy method is built from three layers: Layer 1 (the spec), Layer 2 (the verifier), and Layer 3 (the environment); these three layers together form an end-to-end system for fast, high-quality AI-assisted building.Stop Prompting Claude. Use Karpathy's Method Instead.
- When AI creates a spec for you, you must use your brain to think critically about what the spec says; instructing Claude to 'Make me verify key decisions explicitly to ensure nothing is missed' helps enforce critical human review of the spec itself.Stop Prompting Claude. Use Karpathy's Method Instead.