# Vercel — Eval-driven development: Build better AI faster

- Company: Vercel (vercel.com)
- Announced: 2024-10-17T13:00:00+00:00
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://vercel.com/blog/eval-driven-development-build-better-ai-faster
- Record: https://forck.live/items/20935-eval-driven-development-build-better-ai-faster
- Subject: v0 / AI SDK / AI Gateway

Vercel describes its philosophy of eval-driven development for AI-native applications, explaining how evaluations (evals) serve as a testing paradigm for probabilistic systems. The article covers three types of evals: code-based grading, human grading, and LLM-based grading, and illustrates their use with an example of generating a React component.

## Evidence

Verbatim from https://vercel.com/blog/eval-driven-development-build-better-ai-faster:

> Evaluations (evals) are like end-to-end tests for AI and other probabilistic systems. They assess output quality against defined criteria using automated checks, human judgment, and AI-assisted grading.

---

Record: https://forck.live/items/20935-eval-driven-development-build-better-ai-faster
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md