Design a News-Filtering Prompt
Company: Point72
Role: Machine Learning Engineer
Category: Machine Learning
Difficulty: medium
Interview Round: Technical Screen
You are acting as the coach of an Olympic champion. The athlete receives many news articles every day, and you want to use a large language model to filter the news so that only useful, relevant, and actionable items are shown to the athlete and coaching team.
Design a prompt for the language model that screens incoming news articles. The prompt should:
1. Define the model's role and objective clearly.
2. Specify what kinds of news should be considered relevant to the athlete.
3. Specify what kinds of news should be filtered out.
4. Require the model to produce a structured output, such as a decision, relevance score, category, short summary, and reason.
5. Include safeguards against speculation, sensationalism, irrelevant gossip, and unsupported claims.
Assume each input article contains at least a headline, source, publication date, and article text.
Overview: This question evaluates prompt engineering and LLM application skills, including role and objective specification, relevance criteria definition, structured output design, and implementation of safeguards against speculation and sensationalism.
Read the full Point72 Machine Learning Engineer interview experience this question came from