← Intuit Interview Insights

Intuit·Machine Learning Engineer·Technical Phone Screen·Senior

Senior
Jun 2026

Summary

Interviewed for an MLE role at Intuit and got a solid block of questions around feature stores. Nothing too wild but the offline-online consistency question tripped me up more than I expected.

Questions Asked (4)

Q1

Have you worked with feature stores before? Walk me through your experience.

System DesignTechnical Trade-offs
Author's notes

Pretty open-ended, which I actually appreciated.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Begin by briefly confirming your experience level with feature stores, then walk through a concrete example using the STAR method, focusing on the business problem, your design decisions, and measurable outcomes. Tailor your answer to highlight system design thinking and trade-offs, which are explicitly relevant to this role at Intuit.

Pro tip: Mention specific feature store platforms (e.g., Feast, Tecton, Hopsworks, or a homegrown solution) and articulate the distinction between the online and offline store — this signals deep practical knowledge rather than surface-level familiarity, and will resonate strongly with a senior ML engineering audience.

1. Establish Context & Scope

Briefly state the project or team context where you used a feature store, including the scale (number of features, models, or users served). This grounds your answer in real-world credibility immediately.

2. Describe the Problem It Solved

Explain the pain points that led to adopting or building a feature store, such as training-serving skew, feature duplication across teams, or slow feature engineering pipelines. This shows you understand the 'why' behind the technology.

3. Walk Through Your Role & Design Decisions

Detail your specific contributions — whether you designed the schema, integrated it into the ML pipeline, or evaluated vendor solutions — and highlight any trade-offs you navigated (e.g., latency vs. consistency, batch vs. streaming ingestion).

4. Address Technical Trade-offs

Discuss key architectural decisions such as online vs. offline store separation, point-in-time correctness for training data, or storage backend choices (Redis, DynamoDB, BigQuery). This directly addresses the 'Technical Trade-offs' category of the question.

5. Share Outcomes & Lessons Learned

Quantify the impact where possible (e.g., reduced feature development time by X%, eliminated training-serving skew) and share one honest lesson learned or limitation encountered. This demonstrates maturity and continuous improvement mindset.

Key Points to Mention

  • Online vs. offline store architecture and the importance of point-in-time correct feature retrieval for model training
  • Training-serving skew prevention and how a feature store enforces consistency between training and inference pipelines
  • Feature reusability and governance — enabling multiple teams or models to share and discover features without duplication
  • Ingestion patterns: batch vs. streaming feature computation and the trade-offs in latency, freshness, and infrastructure cost
  • Specific tools or platforms used (e.g., Feast, Tecton, Hopsworks, SageMaker Feature Store) and why they were chosen over alternatives
  • Monitoring and data quality — how you ensured feature freshness, detected drift, and maintained SLAs for real-time serving

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

What feature store technologies have you used, and what were your reasons for choosing them?

System DesignTechnical Trade-offs
Author's notes

Named a few tools I'd worked with.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Ground your answer in concrete hands-on experience with specific feature store technologies, then articulate the trade-offs that drove your selection decisions rather than just listing tools. Tie your choices back to business or engineering requirements such as latency, scalability, consistency, and team maturity to demonstrate systems-thinking depth relevant to Intuit's scale.

Pro tip: Interviewers at product-driven companies like Intuit care less about which tool you picked and more about whether you can reason through build-vs-buy decisions and operational costs — explicitly mentioning how you evaluated online vs. offline serving consistency or point-in-time correctness will signal senior-level thinking.

1. Name the Technologies

Briefly list the feature store technologies you have directly worked with (e.g., Feast, Tecton, Hopsworks, AWS SageMaker Feature Store, Vertex AI Feature Store, or a custom-built solution). Be honest about the depth of your experience with each.

2. Set the Context

Describe the project or system context in which you evaluated or used these tools — including team size, data volume, model serving latency requirements, and existing infrastructure — so the interviewer understands the constraints that shaped your decision.

3. Explain the Selection Criteria

Walk through the key technical and organizational criteria you used to evaluate options, such as online/offline serving consistency, point-in-time correctness, latency SLAs, storage backend flexibility, managed vs. self-hosted trade-offs, and cost.

4. Articulate the Trade-offs

Honestly discuss what you gained and what you gave up with your chosen solution — for example, Feast offers flexibility but requires more operational overhead, while Tecton is managed but adds vendor dependency and cost.

5. Reflect on Outcomes and Lessons

Share measurable outcomes where possible (e.g., reduced feature pipeline duplication, improved model refresh latency) and mention what you would do differently or evaluate next time, demonstrating continuous learning.

Key Points to Mention

  • Online vs. offline store architecture and the importance of low-latency serving for real-time inference use cases like fraud detection or personalization
  • Point-in-time correctness to prevent training-serving skew, especially critical in financial ML contexts like Intuit's tax or payments products
  • Trade-offs between managed services (Tecton, SageMaker Feature Store) and open-source self-hosted solutions (Feast) in terms of operational burden, cost, and flexibility
  • Integration with existing data infrastructure such as Spark, Kafka, Snowflake, or BigQuery and how that influenced the choice
  • Feature reuse and discoverability — how a feature store reduces duplication across teams and accelerates model development
  • Monitoring and data quality concerns such as feature drift detection and freshness guarantees in a production environment

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

What do you see as the core responsibilities of a feature store in an ML system?

System DesignData Modeling
Author's notes

I covered the basics: feature reuse, versioning, serving at low latency.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Frame your answer by first defining what a feature store is and the core problem it solves, then systematically walk through its key responsibilities across the ML lifecycle. Ground your explanation in real-world pain points like training-serving skew and feature reuse to demonstrate practical experience rather than just theoretical knowledge.

Pro tip: Mention the dual nature of a feature store — serving both offline (training) and online (inference) use cases — and explain how bridging these two worlds consistently is arguably its most critical and technically challenging responsibility. This shows you understand the nuanced engineering tradeoffs involved.

1. Define the Core Problem

Start by articulating why a feature store exists — to eliminate redundant feature engineering, prevent training-serving skew, and provide a centralized, governed repository for ML features. This sets the context for all responsibilities you'll discuss.

2. Cover Feature Storage & Serving

Explain the dual storage responsibility: an offline store (e.g., data warehouse or data lake) for batch training data and an online store (e.g., Redis, DynamoDB) for low-latency real-time inference. Emphasize that consistency between these two is paramount.

3. Discuss Feature Computation & Transformation

Describe how a feature store manages feature pipelines — both batch and streaming — to compute, transform, and keep features fresh. Mention point-in-time correctness to avoid data leakage during training.

4. Address Discovery, Reuse & Governance

Highlight the feature registry and catalog responsibilities: enabling teams to discover existing features, understand their lineage, and reuse them across models. Include data quality monitoring, access control, and versioning as governance pillars.

5. Connect to Business Impact

Tie these responsibilities back to tangible outcomes — faster model development cycles, reduced engineering duplication, and more reliable production models. For a company like Intuit, you could reference financial data features shared across fraud detection and personalization models.

Key Points to Mention

  • Training-serving skew prevention — ensuring features used in training exactly match those served at inference time
  • Point-in-time correctness — retrieving historical feature values as they existed at a specific timestamp to prevent data leakage
  • Offline vs. online store architecture — batch storage for training (e.g., S3, BigQuery) vs. low-latency online storage for inference (e.g., Redis)
  • Feature registry and discoverability — a centralized catalog with metadata, ownership, and lineage for cross-team reuse
  • Feature freshness and pipeline orchestration — managing batch and streaming ingestion to keep features up to date
  • Versioning and governance — tracking feature versions, enforcing access controls, and monitoring data quality and drift

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

How do you make sure features computed offline match what gets served online at inference time?

System DesignTechnical Trade-offsRoot Cause Analysis
Author's notes

This is where I stumbled.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Frame your answer around the concept of 'training-serving skew' and explain it as a systems problem requiring both preventive design and detective monitoring. Walk through the architectural decisions that enforce consistency at each stage of the feature pipeline, then discuss how you validate and alert on drift in production. Ground your answer in concrete tools and patterns you've used or would use at scale.

Pro tip: Mention that skew is often silent — models degrade gradually without hard errors — so proactive monitoring and shadow-mode comparisons are more valuable than reactive debugging. Bringing up a real example of skew you've caught (or a near-miss) signals hands-on experience and maturity.

1. Define the Problem (Training-Serving Skew)

Briefly explain that training-serving skew occurs when the feature distribution or computation logic differs between offline training and online inference, silently degrading model performance. Set the stage by acknowledging this is one of the most common and costly production ML issues.

2. Enforce Shared Feature Logic

Describe using a centralized feature store (e.g., Feast, Tecton, or an internal platform) or shared transformation libraries so that the exact same code path computes features offline and online. Emphasize that duplicating logic in two codebases is the root cause of most skew.

3. Validate at Training and Serving Boundaries

Explain adding schema validation and statistical checks (e.g., Great Expectations, custom assertions) at both the point of feature materialization and at inference time to catch type mismatches, null rates, or distribution shifts early. Include point-in-time correctness checks for time-sensitive features to prevent data leakage.

4. Monitor and Alert in Production

Discuss logging a sample of online feature vectors and comparing their distributions against the training baseline using statistical tests (KS test, PSI) or purpose-built tools like Evidently or WhyLogs. Set up automated alerts so skew is caught before it materially impacts business metrics.

5. Establish a Shadow Mode / Canary Testing Practice

Before full rollout, run the online feature pipeline in shadow mode alongside the offline pipeline and diff the outputs on the same inputs to surface any discrepancies. Treat this as a required gate in the deployment checklist, not an optional step.

Key Points to Mention

  • Feature store as a single source of truth to eliminate duplicated transformation logic between training and serving
  • Point-in-time correctness — ensuring offline features are computed with only data available at the prediction timestamp to avoid leakage
  • Schema and statistical validation at pipeline boundaries using tools like Great Expectations or custom checks
  • Production monitoring of feature distributions using metrics like Population Stability Index (PSI) or KS-test with alerting thresholds
  • Shadow mode / offline-online diff testing as a pre-deployment gate to catch skew before it reaches users
  • Versioning of feature transformations alongside model versions so rollbacks are consistent across both artifacts

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.