# 23andMe Interview Guide (2026): Process, Questions, Compensation

Source: https://www.techinterview.org/companies/23andme-interview-guide/
Updated: 2026-07-12 · techinterview.org

**TL;DR —** 23andMe's interview process usually moves from a recruiter phone screen to a technical or hiring-manager round and then a multi-stage onsite loop, with the exact stages depending on whether you're interviewing for engineering, data science, bioinformatics, or product. Expect role-specific technical questions paired with behavioral rounds that probe how you collaborate across scientific and consumer-facing teams. Compensation typically combines base salary, equity, and a performance bonus, in line with other mid-size Bay Area biotech and tech employers.

23andMe Interview Guide

**Company overview:** 23andMe is a direct-to-consumer genetic testing company that pioneered the consumer genomics market, offering ancestry and health-related genetic reports. Sunnyvale, California headquartered. The company underwent significant financial restructuring in 2024–2025, including bankruptcy proceedings and reorganization; engineering hiring has been reduced from peak years but is ongoing for selected roles in the surviving entity. Hiring candidates should verify the company's current state given the recent corporate changes.

## The post-restructuring context

Through the 2010s, 23andMe was a high-flying consumer biotech with peak hiring across genomics, ML, drug discovery (in partnership with major pharma), and consumer product engineering. The 2024–2025 restructuring included substantial layoffs, divestitures, and a refocus on the core consumer genomics business plus a smaller research arm. As of 2026, engineering hiring is selective and focused on maintaining the core platform; the explosive-growth era is over. Candidates should verify role stability and company runway as part of [evaluating an offer](https://www.techinterview.org/post/3233474669/salary-negotiation-2026/).

## Interview process

Timeline: 4–6 weeks.

- **Recruiter screen.** A short call covering your background, why 23andMe, and logistics like location and pay expectations. Given the restructuring, expect the recruiter to be direct about role scope and to gauge whether you are comfortable joining a company in a leaner phase.

- **Hiring manager screen.** The manager digs into your recent work and whether your experience fits the specific team's needs. Come with a clear story of what you personally owned on past projects, and bring questions about the team's roadmap and stability.

- **Technical phone screen (60 min).** One or two coding problems at medium difficulty in a shared editor. Talk through your approach before you write code, and state the [time and space complexity](https://www.techinterview.org/big-o-cheat-sheet/) of your solution.

- **Virtual onsite (4 rounds).**

    1–2 coding rounds (medium difficulty) — expect array, string, hash-map, and graph-traversal problems; clean, working code with edge cases handled matters more than clever tricks. Test your solution on a small example before you say you are done.

    - 1 [system design round](https://www.techinterview.org/category/system-design/) — you may design a data pipeline or a customer-facing service rather than a generic social app. Focus on the data model, throughput, and how you would move and store sensitive genetic data safely.

    - 1 behavioral round, sometimes including questions about adapting to organizational changes — expect questions on collaboration, ownership, and ambiguity, plus at least one about working through the restructuring. Use concrete examples with your specific actions and the outcome; the [STAR method](https://www.techinterview.org/post/3233460379/behavioral-interview-questions-2026-star-method-amazon-leadership-principles-and-winning-answers/) keeps those stories tight.

- **Hiring committee review.** A group reads the written feedback from every round and makes the final call, so consistency across interviewers matters. A single weak signal can stall an otherwise strong loop.

## Common technical questions

- [Standard LeetCode mediums](https://www.techinterview.org/problems-by-difficulty/) — arrays, strings, hash maps, two pointers, and graph traversal show up most. Practice explaining the trade-offs out loud, not just reaching an answer.

- Genomics pipelines: imputation, ancestry inference, GWAS, polygenic scoring. For genomics-track roles, be ready to explain how these steps connect — imputation fills in ungenotyped variants, GWAS finds trait-associated SNPs, and polygenic scoring aggregates those into a single risk estimate. You do not need a PhD, but you should reason clearly about data scale and statistical confidence.

- Privacy-preserving data practices: differential privacy, secure multi-party computation, PII handling at scale. Interviewers probe whether you can protect genetic data while keeping it useful for research. Be ready to discuss what noise a differential-privacy budget adds and why raw genotypes should never cross a controlled boundary.

- For consumer-facing engineering: large-scale React applications, mobile considerations. Expect questions on rendering large result sets without freezing the browser and on keeping report data responsive on slower devices. Component structure, state management, and lazy loading are common discussion points.

- For ML-track: applying ML to genomic and phenotypic data, evaluation of clinical-grade predictions. The bar here is evaluation rigor: how you validate a model whose output informs a health decision, how you handle class imbalance for rare conditions, and how you avoid overfitting to a non-representative population.

## Compensation (2026 estimates)

[Compensation](https://www.techinterview.org/total-comp-calculator/) is below pre-restructuring levels. New offers are cash-heavy with limited equity given the restructured cap table.

- **Mid:** $140–180K base + bonus → $160–200K total

- **Senior:** $180–240K base + bonus → $210–280K total

- **Staff:** $240–300K base → $270–340K total

## Sample interview questions in depth

### Coding (Python / Go)

- **Compute polygenic risk scores at scale.** Given a million genotyped customers and a PRS model with thousands of weighted SNPs, compute scores efficiently. Discuss vectorization with NumPy/PyTorch, how to handle missing genotypes, and how to roll out new score versions without recomputing everything.

- **Implement an ancestry inference pipeline.** Local ancestry inference (which segments of a chromosome come from which population). Discuss reference panels, the role of statistical models like Hidden Markov Models, and the tension between resolution and confidence.

- **Design a relative-finder.** Identify likely cousins/half-siblings/etc among the user base by detecting shared chromosomal segments (IBD — identity by descent). Discuss the privacy implications and how the matching is opt-in.

### Privacy-preserving practices (senior+)

- Differential privacy applied to research releases: epsilon budget, the privacy-utility trade-off, and why clinical conclusions can still be drawn from properly noised aggregate data.

- Multi-party computation for cross-database queries: how research collaborations work without raw data sharing.

- The 2023 data breach context: what changed in 23andMe's privacy posture (credential stuffing detection, opt-in defaults for relative-finder, audit logging). Engineering questions sometimes reference specific lessons from that incident.

### Consumer product engineering

- React/Next.js for the customer-facing genome browser: how to render arbitrary chromosome regions efficiently, how to handle very large data sets in the browser without freezing the UI.

- Mobile considerations: the iOS/Android apps for accessing reports, how to balance offline access against the need to gate access on session validity.

- A/B testing on health-information features: ethical constraints (you cannot A/B test the accuracy of a health report), regulatory constraints (FDA views some changes as device modifications).

## The post-restructuring landscape

23andMe's 2024-2025 financial restructuring reduced the company's scale significantly. The remaining engineering organization is leaner, more focused on the consumer genomics core, and operates with tighter budgets than at peak. Candidates should evaluate the company's runway as part of any offer decision and ask about specific team stability during the loop.

## What survives is the data moat

Even at reduced scale, 23andMe holds one of the largest consumer genetic databases in the world. Engineering work that depends on that data scale (large-scale GWAS, ancestry refinement, population-genetics research) continues to be high-leverage. New roles in 2026 are concentrated on extracting more value from existing data rather than expanding to new product surface area.

## Frequently Asked Questions

### Is 23andMe a stable employer in 2026?

The post-restructuring entity is operating but has reduced scale significantly. Candidates should evaluate company runway and role stability carefully as part of an offer decision. Recent news on the company's financial state is essential reading.

### Do I need genomics background?

For genomics-track roles yes. For consumer-product engineering, no.

### What is the privacy posture?

23andMe holds extremely sensitive genetic data. Privacy-engineering practices are central to the engineering culture. Candidates joining should expect heavy emphasis on data-handling discipline.

### What happened in the restructuring?

Years of losses combined with a large 2023 data breach contributed to financial stress. The company restructured through 2024–2025 with layoffs and refocusing on core consumer genomics. Specific corporate-state details have continued to evolve; verify current status before pursuing an offer.

## Adjacent Healthcare Tech

- [Tempus AI](https://www.techinterview.org/companies/tempus-ai-interview-guide/) — precision medicine and genomics