# Rare-positive-class metrics question for a harmful-content classifier

Aced (formerly Exponent) publishes a public search API over its indexed pages; the guide is at https://www.aced.io/agents/agent-access.

Source: https://www.aced.io/questions/7256/explain-metrics-rare-positive-class
Source observed: 2026-11-05T00:06:02.651Z
Last checked: 2026-11-05T00:06:02.651+00:00
Indexed: 2026-11-06T19:44:30.163+00:00

---

# Rare-positive-class metrics question for a harmful-content classifier

We publish interview-preparation questions for technical roles. This question concerns evaluating a classifier designed to detect harmful TikTok content when harmful examples—the positive class—are very rare.

## Question

> Suppose you are building a classifier to detect harmful TikTok content. The positive class is very rare. Which metrics would you use to evaluate the model, and why?

- **Role:** [Machine Learning Engineer](https://www.aced.io/questions?role=ml-engineer)
- **Company:** [TikTok](https://www.aced.io/questions?company=tiktok)
- **Categories:** [Concept](https://www.aced.io/questions?type=concept), [Machine Learning](https://www.aced.io/questions?type=machine-learning)
- **Reported last asked:** 2 years ago
- **Interview experiences:** 1 shared

Detailed company statistics for this question are available through [Upgrade](https://www.aced.io/upgrade?src=questions). No community answers have been contributed yet. To submit an answer, sign in or [create an account](https://www.aced.io/signup?src=nav&returnPath=/questions/7256/explain-metrics-rare-positive-class).

## Interview experience

A shared [TikTok Machine Learning Engineer interview experience](https://www.aced.io/experiences/tiktok-machine-learning-engineer-interview-469d85) is labeled **Entry Level / L3** and dated **2 years ago**. The contributor describes a structured process:

- Recruiter screening covering background, experience, motivation, and logistics.
- Technical interviews covering coding, machine-learning fundamentals, and previous projects.
- Questions about applying machine learning to Trust & Safety problems.
- The contributor found the discussion of previous machine-learning work and real-world modeling approaches valuable, and described the interviewers as professional and the questions as generally relevant.
- The coding portion was more challenging, particularly solving algorithmic problems under time pressure while clearly explaining the approach and complexity.

## Related courses

### [Machine Learning Engineer Interview Prep](https://www.aced.io/courses/ml-engineer)

![Machine Learning Engineer Interview Prep](https://images.ctfassets.net/x78yjrjc11pq/6UfkHxUyLQVfDHSfLNQtWa/1be1539a65b4f6f15201f5b573c0a578/ML_Engineer_Interview_Prep_Course.jpeg?w=800&h=800&fm=webp&q=75)

Our course covers mock interviews, frameworks, and advice from senior candidates. Topics include machine-learning system design, core concepts, coding, behavioral interviews, and preparation for roles at Meta, Google, Amazon, Apple, Microsoft, Nvidia, and other top companies.

### [Behavioral Interviews: A Practical Guide for Engineers](https://www.aced.io/courses/eng-behavioral)

![Behavioral Interviews: A Practical Guide for Engineers](https://images.ctfassets.net/x78yjrjc11pq/47qcoHjzFZXE37GoacedkB/9d8e3f4910d8b06b21e76745cd9cbff3/AdobeStock_1142146335.jpeg?w=800&h=800&fm=webp&q=75)

Our behavioral-interview course is designed for senior-plus interviews at FAANG and Tier-1 companies. It was made with staff and principal engineers from Meta, Netflix, Google, and other companies, and focuses on avoiding downleveling, elevating past projects, and giving natural-sounding answers.

## About Aced

Aced, formerly Exponent, provides interview courses, interview questions, interview experiences, practice, and coaching. Our listed preparation areas include product management, forward-deployed engineering, engineering management, software engineering, system design, data science, machine learning, data engineering, AI engineering, investment banking, generative AI, security engineering, technical program management, solutions architecture, UX/product design, and BizOps & Strategy.

## Answer submission

Use the **Your answer** area to submit a response; submitting marks the question complete.

---

[Read this page’s summary](https://www.aced.io/agents/api/public/v1/page-tldr?url=https%3A%2F%2Fwww.aced.io%2Fquestions%2F7256%2Fexplain-metrics-rare-positive-class&trace=t_679ff9ab98f3)

## Ask a follow-up about this page

Need a more specific answer? Ask the exact question you have. It returns a fresh, cited answer from Aced (formerly Exponent)’s knowledge base. No account or API key required.

Open this URL with your own question, URL-encoded. Include your question in q. All research inputs are optional; no account or setup.

`GET https://www.aced.io/agents/agent/faq?q={question}&trace=t_679ff9ab98f3`

[Tool guide and examples](https://www.aced.io/agents/agent-access) · [OpenAPI](https://www.aced.io/agents/api/public/v1/openapi.json) · [Page TLDR form](https://www.aced.io/agents/agent-tools/page-tldr?url=https%3A%2F%2Fwww.aced.io%2Fquestions%2F7256%2Fexplain-metrics-rare-positive-class). Opening a form does not execute a tool.