← Basis Ai Interview Insights

Basis Ai·Software Engineer·Onsite - System Design / Architecture·Senior

SeniorPrefer not to say
May 2026Remote

Summary

System design round at Basis AI for a software engineering role, paired with a PM interviewer which I did not expect. The question was deliberately underspecified and that threw me more than I'd like to admit.

Questions Asked (7)

Q1

Design a mobile app that recommends activities to travelers during an airport layover, given that the interviewer provides almost no requirements upfront.

Product Sense & IdeationAdaptability & AmbiguitySystem Design
Author's notes

I jumped straight into components and the interviewer just watched me flounder for a bit before nudging me.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the problem space and stating your assumptions about the target user, layover duration, and airport context. Then propose a high-level solution that focuses on personalization and real-time constraints, and finally dive into the technical architecture and data sources needed.

Pro tip: Treat the lack of requirements as an opportunity to demonstrate product thinking: explicitly call out the trade-offs between personalization, timeliness, and user effort, and suggest a phased rollout starting with a simple MVP.

1. Clarify and Define Scope

Ask clarifying questions to understand the user, layover scenarios, and business goals. State assumptions about layover duration, airport size, and user preferences.

2. Identify User Needs and Pain Points

Brainstorm the traveler's needs during a layover: time constraints, uncertainty, desire for relaxation or productivity, and local exploration. Prioritize the most impactful needs.

3. Propose a High-Level Solution

Outline the core features: personalized recommendations based on layover time, location, and preferences; real-time updates; and easy booking or navigation. Focus on the MVP.

4. Design the System Architecture

Describe the technical components: data sources (flight APIs, airport maps, user profiles), recommendation engine, and mobile frontend. Discuss scalability and latency considerations.

5. Address Trade-offs and Metrics

Discuss trade-offs between personalization and privacy, and between comprehensive data and development effort. Define success metrics like user engagement and satisfaction.

Key Points to Mention

  • Personalization based on user preferences, layover duration, and time of day
  • Real-time data integration (flight status, security wait times, walking distances)
  • Context-aware recommendations (e.g., near gate, quiet spots, power outlets)
  • Offline functionality and low-latency performance for travelers with poor connectivity
  • Privacy considerations and user consent for location and preference data
  • Scalability to handle peak travel times and multiple airports

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

Q2

What is the core constraint that makes layover recommendations uniquely dangerous, and how should it shape the system's design?

Product Sense & IdeationTechnical Trade-offsSystem Design
Author's notes

This one I actually got right.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Identify the core constraint as the asymmetry of consequences: a bad layover recommendation can strand a traveler in a dangerous or expensive situation, unlike a bad movie recommendation. Then explain how this should drive a design that prioritizes safety, transparency, and fallback options over pure optimization.

Pro tip: Frame the constraint in terms of 'reversibility' and 'blast radius'—layover mistakes are hard to undo and affect real-world safety, so the system should default to conservative, explainable choices and always provide a clear escape hatch.

1. Define the core constraint

State that the unique danger is the high cost of failure: a bad layover can lead to missed connections, unsafe locations, or financial loss, and the user may have no easy recourse.

2. Contrast with typical recommendation systems

Explain that unlike content or product recommendations, layover suggestions have real-world, time-sensitive, and safety-critical consequences, so standard metrics like click-through rate are insufficient.

3. Derive design principles

Propose principles such as safety-first ranking, explainability, user control, and graceful degradation (e.g., always show alternatives and allow easy reversal).

4. Translate into system design

Describe concrete design choices: conservative default parameters, human-in-the-loop for edge cases, robust fallback logic, and clear communication of risks and trade-offs.

5. Validate and iterate

Suggest measuring success by user trust and safety outcomes (e.g., reduced incidents) rather than engagement, and building feedback loops to continuously refine constraints.

Key Points to Mention

  • Asymmetry of consequences: a bad layover can be catastrophic, while a good one is merely convenient.
  • Reversibility: layover decisions are often irreversible once travel begins, so the system must minimize the need for reversal.
  • Blast radius: the impact extends beyond the user to companions, business obligations, and safety.
  • Explainability: users need to understand why a layover is recommended to trust and act on it.
  • Fallback and escape hatches: always provide alternative options and clear paths to change plans.
  • Safety and risk metrics: optimize for minimizing harm, not just for time or cost savings.

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

Q3

Walk through the recommendation pipeline from candidate generation to what the user actually sees.

System DesignTechnical Trade-offsProduct Sense & Ideation
Author's notes

Described it as three stages: generate candidates from the airport content store, run a hard feasibility filter using the layover time budget, then rank what's left.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the recommendation pipeline as a multi-stage system, then walk through each stage from candidate generation to final ranking and presentation. Emphasize the trade-offs at each stage, such as recall vs. precision, latency vs. quality, and how user feedback loops inform the system. Conclude by connecting the pipeline to the user experience and business metrics.

Pro tip: Show awareness of the cold-start problem and how Basis Ai might handle it, and discuss how you would measure the pipeline's success beyond offline metrics, such as online A/B tests and user engagement.

1. Candidate Generation

Explain how you retrieve a large set of potentially relevant items from millions, using techniques like collaborative filtering, content-based filtering, or approximate nearest neighbor search. Highlight the trade-off between recall and computational efficiency.

2. Filtering and Scoring

Describe how you filter out irrelevant or low-quality candidates and score the remaining ones using a more complex model, such as a deep neural network. Discuss how you balance latency and accuracy.

3. Ranking and Re-ranking

Explain how you rank the scored candidates and apply re-ranking rules to incorporate business logic, diversity, or freshness. Mention how you handle position bias and fairness.

4. Final Presentation

Discuss how the ranked list is presented to the user, including UI considerations, pagination, and how you log user interactions for feedback. Emphasize the importance of a seamless user experience.

5. Feedback Loop and Iteration

Explain how user interactions (clicks, purchases, etc.) are fed back into the system to retrain models and improve future recommendations. Mention online evaluation and A/B testing.

Key Points to Mention

  • Two-tower models for candidate generation and embedding-based retrieval
  • Trade-offs between recall and precision at different stages
  • Latency constraints and how to optimize for real-time serving
  • Cold-start problem and solutions like content-based features or exploration
  • Diversity and fairness in ranking to avoid filter bubbles
  • Metrics: offline (NDCG, recall@k) and online (CTR, conversion rate, user engagement)

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

Q4

A user's flight gets delayed by 90 minutes while they are mid-activity. How does the system detect this and update its recommendations in real time?

System DesignAPI & IntegrationsAdaptability & Ambiguity
Author's notes

Honestly the most fun part of the interview.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the system's scope and data sources, then walk through the event-driven pipeline from detection to recommendation update, and finally address real-time delivery and user experience. Emphasize idempotency, latency, and graceful degradation.

Pro tip: Proactively discuss how you'd handle duplicate or out-of-order flight status events, as this shows production maturity and prevents incorrect recommendations.

1. Clarify Requirements and Assumptions

Ask about the system's boundaries: which flight data providers are used, what 'mid-activity' means (e.g., browsing, booking), and what real-time latency is acceptable. State your assumptions explicitly.

2. Detection via Event-Driven Ingestion

Explain how flight status updates are ingested from external APIs or webhooks, normalized, and published to an event stream (e.g., Kafka). Mention polling vs. push and how to detect a 90-minute delay.

3. Real-Time Processing and Recommendation Update

Describe a stream processor that consumes delay events, identifies affected users and activities, and triggers recommendation recomputation. Highlight idempotency and handling out-of-order events.

4. Delivery to the User

Explain how updated recommendations reach the user in real time via WebSockets, SSE, or push notifications, and how to ensure consistency and avoid flickering or stale UI.

5. Reliability, Scalability, and Fallbacks

Discuss monitoring, retries, dead-letter queues, and fallback to periodic polling if the stream fails. Mention how to scale for many concurrent users and flights.

Key Points to Mention

  • Event-driven architecture with message queues (e.g., Kafka, Pub/Sub) for decoupling and scalability
  • Idempotent processing and deduplication to handle duplicate or out-of-order flight status events
  • Real-time delivery mechanisms like WebSockets or Server-Sent Events for instant UI updates
  • Caching and precomputation of recommendations to reduce latency during updates
  • Graceful degradation and fallback strategies (e.g., polling) when real-time streams are unavailable
  • Monitoring and alerting on event lag, processing errors, and end-to-end latency

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

Q5

Your walking-time estimates turn out to be systematically too optimistic and users report nearly missing flights. How do you detect this from telemetry and fix the model?

Product Analytics & MetricsRoot Cause AnalysisA/B Testing & Experimentation
Author's notes

I said look at cases where users arrived at the gate within a few minutes of boarding close, cross-reference with what the app recommended, and see if the estimated walk time was shorter than actual.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining the problem and outlining how to detect systematic optimism in walking-time estimates using telemetry data, such as comparing estimated vs. actual arrival times and analyzing user feedback. Then describe a structured approach to diagnose the root cause, fix the model, and validate the improvement through experimentation.

Pro tip: Emphasize the importance of setting up a feedback loop with real user data and A/B testing to continuously monitor and improve model accuracy, rather than relying solely on offline metrics.

1. Detect the Issue

Use telemetry to compare estimated walking times with actual arrival times, and analyze user reports of near-missed flights to identify systematic bias.

2. Diagnose Root Cause

Investigate potential causes such as outdated assumptions, lack of real-time factors (e.g., weather, construction), or biased training data.

3. Fix the Model

Incorporate real-time data sources, adjust for user-specific walking speeds, and retrain the model with more accurate data.

4. Validate and Iterate

Run A/B tests to measure the impact of the updated model on user outcomes, and set up continuous monitoring for ongoing improvement.

Key Points to Mention

  • Define clear metrics for detection: mean absolute error (MAE), bias (average difference between estimated and actual), and user-reported near-miss rate.
  • Use telemetry data such as GPS traces, timestamps, and user feedback to quantify the problem.
  • Consider external factors like time of day, weather, and airport layout that affect walking speed.
  • Apply A/B testing to compare the old and new models, ensuring statistical significance.
  • Implement a feedback loop for continuous learning and model updates.
  • Communicate the impact in terms of user experience and business metrics (e.g., reduced missed flights).

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

Q6

A merchant pays for placement in the recommendations. How do you design the ranking so paid placement can influence results without ever surfacing an unsafe or irrelevant option, and how do you prove to the business it isn't hurting trust?

Pricing & MonetizationProduct StrategyTechnical Trade-offs
Author's notes

Kept it simple: paid boost only applies after the feasibility filter and a relevance floor, so money can move things up within the safe, relevant set but can't rescue something that fails either check.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the problem as a constrained optimization: paid placement can adjust ranking scores, but only within a safe candidate set that has passed relevance and safety filters. Then describe a two-stage architecture: first, a hard filter for safety and relevance, and second, a ranking model that incorporates a paid boost as a feature. Finally, outline an experimentation plan to measure trust impact and business metrics.

Pro tip: Emphasize that paid placement should never override safety or relevance thresholds; instead, it should only reorder items that already meet quality bars. Propose a transparent audit trail and a kill switch to disable paid boosts if trust metrics degrade.

1. Define safety and relevance gates

Establish hard filters that any item must pass before entering the ranking pool, such as policy violations, unsafe content, or low relevance scores. These gates are non-negotiable and cannot be bypassed by paid placement.

2. Incorporate paid boost as a ranking feature

Within the safe candidate set, add a paid boost signal to the ranking model, but cap its influence to ensure it only reorders items within a quality band. Use a multiplicative or additive boost with a maximum delta.

3. Monitor and mitigate trust impact

Set up A/B tests and guardrail metrics (e.g., user satisfaction, CTR, long-term retention) to detect any negative impact. If trust metrics drop beyond a threshold, automatically reduce or disable paid boosts.

4. Provide business transparency and control

Offer dashboards showing how paid placement affects ranking and trust metrics, and allow business stakeholders to adjust boost parameters within safe limits. Document the trade-offs and decision logic.

Key Points to Mention

  • Two-stage ranking: hard filters for safety/relevance, then soft ranking with paid boost.
  • Paid boost should be a feature in the ranking model, not a hard override.
  • Cap the boost to prevent low-quality items from outranking high-quality ones.
  • Use guardrail metrics (e.g., user trust, satisfaction) in experiments to measure impact.
  • Implement a kill switch or automatic rollback if trust metrics degrade.
  • Maintain an audit trail and transparency for business stakeholders.

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

Q7

How does your feasibility model change for a long international layover where the traveler must re-clear immigration and security to leave the terminal?

Product Sense & IdeationTechnical Trade-offsAdaptability & Ambiguity
Author's notes

Added immigration and security re-entry time as explicit cost components in the time budget, and set a minimum layover threshold before the app even surfaces outside-airport options.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the scenario and the feasibility model's purpose, then systematically identify the new constraints introduced by re-clearing immigration and security. Explain how you would adjust the model's inputs, logic, and outputs to account for these constraints, and discuss trade-offs and validation.

Pro tip: Emphasize that re-clearing immigration and security adds non-deterministic time buffers and potential failure points; propose modeling these as probabilistic distributions rather than fixed times to reflect real-world variability.

1. Clarify the scenario and model objectives

Ask questions to understand the traveler's nationality, visa requirements, layover duration, and the model's current assumptions. Confirm what 'feasibility' means (e.g., making the connection, leaving the airport).

2. Identify new constraints and variables

List factors like immigration processing time, security queue time, visa/entry requirements, terminal transfer time, and re-boarding cutoff. Note which are deterministic vs. stochastic.

3. Adjust model inputs and logic

Incorporate probabilistic distributions for immigration and security times, add buffers for uncertainty, and include conditional logic for visa eligibility. Update the feasibility calculation to sum these times and compare against layover duration.

4. Evaluate trade-offs and edge cases

Discuss how to handle worst-case scenarios (e.g., long queues, denied entry) and trade-offs between model accuracy and simplicity. Consider fallback options like staying airside.

5. Validate and iterate

Propose using historical data or simulations to calibrate the model, and suggest monitoring real outcomes to refine estimates over time.

Key Points to Mention

  • Immigration and security processing times are variable and should be modeled with distributions (e.g., P50, P90) rather than fixed values.
  • Visa and entry requirements can make leaving the terminal infeasible regardless of time; the model must include a binary eligibility check.
  • Re-boarding cutoff times (e.g., 30-45 minutes before departure) reduce the effective layover duration available for leaving the terminal.
  • Terminal transfer time (walking, train, bus) adds to the total time needed and can vary by airport.
  • The model should output a probability of feasibility, not just a binary yes/no, to reflect uncertainty.
  • Trade-offs exist between model complexity (e.g., simulating queues) and practical usability; consider a tiered approach based on data availability.

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