Study plan
Every objective in the published exam guide, grouped by exam section and ordered the way the guide orders them.
0 answered correctly0 not correct, still open56 not tried yet0 of 56 objectives tried
This is a record of what you have tried, not a score and not a prediction of the real exam.
Start hereStudy noteDesign Applications
Writing a prompt that reliably produces a specific response format
Objective 1 of 6 in Design Applications, the first section in the guide. Not tried yet. See all 6 exam sections
What this note covers
Designing prompts that elicit specifically formatted responses (e.g., JSON, XML, Markdown tables, or bullet lists) is essential for building GenAI applications that pipe LLM outputs into downstream parsing, APIs, or UI components. This requires combining clear instructions, examples, and structural cues so the model reliably produces machine-readable or consistently structured output.
The facts the exam tests
- Explicitly state the desired output format in the instruction (e.g., 'Respond only in valid JSON with keys x and y') rather than assuming the model will infer it.
- Few-shot examples showing the exact desired input/output format significantly improve consistency, especially for structured formats like JSON or tables.
- Providing a schema, template, or delimiter (e.g., triple backticks, XML tags, or a JSON skeleton) helps constrain the model's output structure and makes parsing more reliable.
- Instructing the model to avoid extra commentary (e.g., 'Return only the JSON object, no explanation') reduces the risk of downstream parsing errors.
- Even with well-designed prompts, LLM outputs can still be malformed, so applications should include validation/parsing logic (e.g., try/except JSON parsing) and, if needed, retry or repair steps rather than relying on the prompt alone.
- System messages (where supported) are a good place to set persistent formatting instructions that apply across a conversation or multiple calls.
An engineer is prototyping an invoice-processing pipeline where a downstream regex parser expects each response as vendor_name|invoice_date|total_amount on a single line. Testing shows that although the extracted values are usually correct, the field order and delimiter vary between calls, breaking the regex. Which prompt design change should the engineer make to enforce the required output format?
The 6 exam sections
Ordered as the guide orders them
Design Applications10.71% of the exam*0 of 6 tried
Data Preparation14.29% of the exam*0 of 8 tried
Application Development23.21% of the exam*0 of 13 tried
Assembling and Deploying Applications26.79% of the exam*0 of 15 tried
Governance7.14% of the exam*0 of 4 tried
Evaluation and Monitoring17.86% of the exam*0 of 10 tried
* Our estimate. Databricks publishes no section weights.
These are independent practice questions, written against this certification's published exam guide. They are not the certification vendor's own questions, and not the real exam.