# @awomanindatascience on Instagram

- **Type:** Video
- **Original URL:** https://www.instagram.com/p/DXAXRULSnTC
- **Gondola URL:** https://gondola.cc/posts/63699900-awomanindatascience-instagram
- **Thumbnail:** https://img.gondola.cc/tr:w-,h-,fo-auto/postThumbnails/78a276bdb2.jpg
- **Posted:** 2026-04-11T21:00:55.000+00:00
- **Account Owner:** Chinar Arora (@awomanindatascience) — https://gondola.cc/awomanindatascience

## Caption

Most developers are using AI to write code.
Very few are using it the right way.

The difference isn’t the tool-
it’s how you think, structure, and validate your work.

Here are 5 practices that turn AI from a shortcut into a real engineering advantage.

1. Spec before you prompt

Don’t start with “build X.”
Start with a clear spec:
intent, constraints, and what “done” looks like.
AI works best when the problem is well-defined.

2. Context engineering > prompt engineering

Output quality depends more on context than wording.
Provide the right information at the right time:
project rules, architecture, and constraints-not full history or noise.

3. Use the Plan → Execute → Verify loop

Don’t treat AI as one-shot.
Break work into steps:
plan the approach, generate code, then review and refine with specific feedback.

4. Testing is the foundation

AI-generated code often looks correct but isn’t.
Define tests first, then generate code to pass them.
This ensures correctness before implementation.

5. Security and review are non-negotiable

AI can introduce real vulnerabilities.
Always review code, validate dependencies, and enforce security practices before production.

#generativeai #womenintech #vibecoding #codingtools

## Stats

- **Views:** 295
- **Likes:** 17
- **Shares:** 0
- **Comments:** 1

## Tags

codingtools, generativeai, womenintech, vibecoding

---
Copyright (c) Gondola