10 Prompt Engineering Examples: Before and After sinc-LLM

By Mario Alexandre · March 27, 2026 · 11 min read

Theory is good. But examples are better. Here are 10 real prompt engineering examples. They show exactly how sinc-LLM's 6-band system turns vague prompts into clear ones. You can see the difference in the output right away.

Each example works the same way. You see the raw prompt first. Then you see what is wrong with it and which bands are missing. Then you see the structured version. Then you see the result.

Example 1: Blog Post Writing

Result: Before = a generic list with no new ideas. After = a focused post with real tools, cited studies, and advice that fits engineering managers.

Example 2: Code Generation

Result: Before = a simple csv.reader with no error handling. After = a production-ready function with validation, logging, and edge case handling.

Example 3: Email Drafting

Result: Before = a generic follow-up that could be for any client. After = a specific email that names the actual proposal and gives a clear next step.

Example 4: Data Analysis

Result: Before = vague comments about "data trends." After = a real analysis with month-over-month calculations, anomaly identification, and clear recommendations tied to the actual numbers.

Example 5: SQL Query

Result: Before = a basic SELECT with GROUP BY and wrong table assumptions. After = a CTE-based query with correct table names, date filtering, refund exclusion, and a window function for the running total.

The Pattern Across All 10 Examples

Review each example for missing task details. Check CONSTRAINTS, DATA, FORMAT, PERSONA and CONTEXT alongside TASK, and note which omissions matter for that example. Do not infer prevalence across all prompts from this collection.

The sinc-LLM framework is based on the Nyquist-Shannon theorem. That theorem explains why this happens:

Each missing band is a missing sample. Each missing sample adds noise. When you are missing 4 or 5 bands, the output barely looks like what you wanted.

Full sinc JSON Example

{
  "formula": "x(t) = \u03a3 x(nT) \u00b7 sinc((t - nT) / T)",
  "T": "specification-axis",
  "fragments": [
    {"n": 0, "t": "PERSONA", "x": "Expert data scientist with 10 years ML experience"},
    {"n": 1, "t": "CONTEXT", "x": "Building a recommendation engine for an e-commerce platform"},
    {"n": 2, "t": "DATA", "x": "Dataset: 2M user interactions, 50K products, sparse matrix"},
    {"n": 3, "t": "CONSTRAINTS", "x": "Must use collaborative filtering. Latency under 100ms. No PII in logs. Python 3.11+. Must handle cold-start users with content-based fallback"},
    {"n": 4, "t": "FORMAT", "x": "Python module with type hints, docstrings, and pytest tests"},
    {"n": 5, "t": "TASK", "x": "Implement the recommendation engine with train/predict/evaluate methods"}
  ]
}

Try It Yourself

Take any prompt you have used lately. Break it into 6 bands. Or paste it into sinc-LLM and let the tool do it for you. Compare the raw and structured outputs. The difference is easy to see, and you can measure it.

Want to go deeper? Read the complete 2026 guide or see the best practices from 275 experiments.

// Keep reading

• What an AI Fallback Path Actually Looks Like in a Production System
• AI Monitoring Checklist: What Every Production System Needs on Every Critical Path
• SLO Design for AI Systems: How to Set Error Budgets That Survive a Vendor Update

// Production AI Engineering

Build AI systems that hold up in production.

sinc-LLM designs, audits, and stabilises production AI infrastructure: from vendor evaluation and cost accountability to incident controls and MCP architecture.

See what we do →