Prompt Engineering 101

Prompt engineering is the skill of communicating effectively with AI models to get the output you want. It's the single most impactful skill for anyone using AI tools today — and it doesn't require any coding knowledge.

Whether you're using ChatGPT, Claude, Gemini, or Copilot, these techniques will dramatically improve your results.

Why Prompts Matter

The same AI model can give you a brilliant answer or a useless one — the difference is almost always in how you ask. A vague prompt gets a vague answer. A specific, well-structured prompt gets a specific, useful answer.

Bad:  "Tell me about AI"
Good: "Explain how Indian BFSI companies are using AI for fraud detection,
       with 3 specific examples. Keep it under 300 words."

The CLEAR Framework

We teach a simple framework for writing effective prompts:

C — Context

Tell the AI who you are and what situation you're in.

"I'm an L&D manager at a mid-size Indian IT services company with 5,000 employees."

L — Length & Format

Specify the output format and length you want.

"Give me a bullet-point summary, no more than 10 points."

E — Examples

Show the AI what good output looks like.

"Format each point like: [Department] — [Use Case] — [Expected Impact]"

A — Action

Be explicit about what you want the AI to do.

"Create a 90-day AI training roadmap for our engineering team."

R — Restrictions

Tell the AI what to avoid.

"Don't include any tools that require a paid subscription. Focus on open-source options."

Essential Techniques

1. Role Assignment

Tell the AI to adopt a specific persona:

"You are a senior data scientist at an Indian bank. Review this ML model
proposal and identify potential risks specific to the Indian regulatory
environment."

This works because it primes the model to draw on knowledge relevant to that role.

2. Chain of Thought

Ask the AI to think step by step:

"Analyse this business problem step by step:
1. First, identify the root cause
2. Then, list possible solutions
3. Evaluate each solution's feasibility in an Indian enterprise context
4. Recommend the best approach with justification"

This produces more thorough, well-reasoned outputs.

3. Few-Shot Examples

Give the AI examples of what you want:

"Rewrite these job descriptions to be more inclusive. Here's an example:

Before: 'Looking for a rockstar developer who can crush deadlines'
After: 'Looking for a skilled developer who delivers quality work on schedule'

Now rewrite these:
1. 'Need an aggressive salesperson who can dominate the market'
2. 'Seeking a young, energetic team player'
"

4. Output Templating

Define the exact structure you want:

"For each AI tool, provide:
- Tool name:
- Category: [Productivity / Coding / Design / Data]
- Free tier: [Yes / No]
- Best for:
- Indian enterprise readiness: [High / Medium / Low]"

5. Iterative Refinement

Don't expect perfection on the first try. Build on the AI's response:

• "Good, but make it more concise"
• "Rewrite point 3 with a specific Indian example"
• "Now convert this into a presentation outline with 10 slides"

Common Mistakes

Being Too Vague

Bad:  "Help me with my presentation"
Good: "Create a 10-slide outline for a presentation to my CEO about
       why we should invest ₹50 lakhs in AI training for our 200-person
       engineering team. Include ROI projections."

Not Providing Context

Bad:  "Write a follow-up email"
Good: "Write a follow-up email to an L&D head at Infosys who attended
       our AI training demo last week. They seemed interested but haven't
       responded. Keep it professional but warm, under 150 words."

Accepting the First Output

The first response is rarely the best. Always iterate. Ask the AI to improve, add detail, change tone, or try a different approach.

Prompts for Common Enterprise Tasks

Meeting Notes

"Summarise these meeting notes into:
1. Key decisions made (bullet points)
2. Action items (who, what, by when)
3. Open questions to resolve
Keep it under 200 words."

Email Drafting

"Draft a professional email to [recipient] about [topic].
Tone: [formal/friendly/urgent]
Key points to include: [list]
Length: Under 150 words"

Data Analysis

"I have quarterly sales data for our 5 Indian regions.
Identify:
1. Top performing region and why
2. Underperforming regions with possible reasons
3. Trends worth investigating
4. Recommendations for next quarter"

Document Review

"Review this vendor proposal and identify:
1. Missing information we should request
2. Potential risks or red flags
3. Comparison points against our requirements
4. Questions for the follow-up call"

Next Steps

Prompt engineering is a skill that improves with practice. Start by:

1. Using the CLEAR framework for your next 10 AI interactions
2. Saving prompts that work well in a personal prompt library
3. Sharing effective prompts with your team
4. Experimenting with different models — what works in ChatGPT may need adjustment for Claude

The difference between an AI novice and an AI power user is not the tool — it's the prompt.