← Atlassian Interview Insights

Atlassian·Machine Learning Engineer·Onsite - System Design / Architecture·Senior

SeniorPrefer not to say
Jul 2026

Summary

Atlassian ML Engineer interview focused almost entirely on designing a recommendation system for a product feed. It was a deep system design session and they really wanted to see you hold the whole pipeline in your head at once, from retrieval to serving to how you'd know if it's actually working.

Questions Asked (5)

Q1

Design a recommendation system for a product feed (like an e-commerce home feed). Walk through the full system including requirements, architecture, and how you'd evaluate it.

System DesignTechnical Trade-offsA/B Testing & Experimentation
Author's notes

This was the whole interview basically.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying requirements and constraints, then propose a high-level architecture that covers data, modeling, serving, and evaluation. Emphasize trade-offs and how you would measure success through offline metrics and online A/B tests.

Pro tip: Atlassian values data-driven decisions and collaboration; tie your design to business metrics like user engagement and revenue, and discuss how you'd iterate with cross-functional teams.

1. Clarify Requirements

Ask questions to understand the product, scale, latency, and business goals. Define functional and non-functional requirements.

2. High-Level Architecture

Outline the end-to-end system: data collection, feature engineering, model training, serving, and feedback loops. Consider batch vs. real-time components.

3. Modeling Approach

Choose recommendation algorithms (e.g., collaborative filtering, content-based, deep learning) and explain how you'd handle cold start, diversity, and freshness.

4. Evaluation Strategy

Define offline metrics (e.g., recall@k, NDCG) and online metrics (e.g., CTR, conversion). Describe A/B testing setup, guardrail metrics, and statistical significance.

5. Trade-offs and Iteration

Discuss trade-offs between different approaches (e.g., latency vs. accuracy, exploration vs. exploitation) and how you'd iterate based on results.

Key Points to Mention

  • Two-stage architecture: candidate generation and ranking
  • Feature store for consistent online/offline features
  • Handling cold start with content-based or hybrid methods
  • Real-time personalization and session-based features
  • Offline evaluation metrics like precision@k, recall@k, NDCG
  • Online A/B testing with guardrail metrics and long-term holdouts

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

Q2

How would you handle cold start for new users and new items in this recommendation system?

System DesignTechnical Trade-offs
Author's notes

They pushed on this specifically after I glossed over it.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging that cold start is a fundamental challenge in recommendation systems, then structure your answer around separate strategies for new users and new items, emphasizing hybrid approaches that combine content-based and collaborative filtering. Conclude by discussing how to measure and iterate on these strategies, and mention trade-offs between exploration and exploitation.

Pro tip: Show maturity by discussing how you would validate cold-start strategies offline and online, and mention the importance of fallback mechanisms to avoid poor user experiences. Also, highlight that cold start is not a one-time problem but requires continuous monitoring and adaptation.

1. Define the cold-start problem

Clearly distinguish between new user cold start (lack of user interaction data) and new item cold start (lack of item interaction data), and explain why they require different approaches.

2. Strategies for new users

Discuss onboarding techniques such as asking for explicit preferences, using demographic or contextual information, and employing popularity-based or diverse recommendations to gather initial signals.

3. Strategies for new items

Explain content-based approaches using item metadata (e.g., tags, descriptions, categories) and how to leverage multi-armed bandits or exploration strategies to quickly gather interaction data.

4. Hybrid and transfer learning approaches

Describe how to combine content-based and collaborative filtering, and how transfer learning from related domains or meta-learning can help when data is scarce.

5. Evaluation and iteration

Outline how to measure the effectiveness of cold-start strategies using offline metrics (e.g., coverage, diversity) and online A/B tests, and how to iterate based on feedback.

Key Points to Mention

  • Content-based filtering using item features and user profiles
  • Exploration-exploitation trade-off and multi-armed bandits
  • Hybrid recommendation systems combining multiple signals
  • Transfer learning and meta-learning for cold start
  • Fallback strategies like popularity or trending items
  • Evaluation metrics for cold start: coverage, novelty, and user engagement

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

Q3

What offline metrics would you use to evaluate the ranker, and how do they connect to what you actually care about in production?

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

Talked through NDCG, precision at k, and AUC.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by outlining standard offline ranking metrics like NDCG, MAP, and MRR, then explicitly connect each to online business metrics such as click-through rate, conversion, and user engagement. Emphasize that offline metrics are proxies and must be validated through online A/B tests, and discuss how to choose metrics that align with the product's goals.

Pro tip: Mention that offline metrics should be computed on a held-out set that mimics production traffic distribution, and that you'd monitor for metric drift and calibration to ensure offline gains translate online.

1. List common offline ranking metrics

Mention metrics like NDCG, MAP, MRR, Precision@k, and Recall@k, explaining what each measures (e.g., NDCG for graded relevance, MAP for precision across relevant items).

2. Map offline metrics to online objectives

Connect each offline metric to production goals: e.g., NDCG correlates with user satisfaction and CTR, MAP with conversion rate, MRR with time-to-first-click.

3. Acknowledge limitations and biases

Discuss how offline metrics can be misleading due to position bias, selection bias, or feedback loops, and that they don't capture long-term effects or diversity.

4. Validate with online experiments

Explain that offline metrics are used for rapid iteration, but final validation requires A/B tests measuring business metrics like revenue, retention, and task success.

5. Iterate and refine metrics

Show how you'd continuously refine offline metrics based on online results, possibly using counterfactual evaluation or inverse propensity scoring to better estimate online performance.

Key Points to Mention

  • NDCG, MAP, MRR, Precision@k, Recall@k
  • Online metrics: CTR, conversion rate, dwell time, task success, retention
  • Position bias and how to correct it (e.g., using unbiased learning to rank)
  • A/B testing as the gold standard for validation
  • Counterfactual evaluation and off-policy evaluation techniques
  • Alignment with product goals: e.g., Atlassian's focus on collaboration and productivity

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

Q4

How would you design the feature store and training pipeline to keep features fresh enough for a real-time feed?

System DesignData Modeling
Author's notes

Honestly the part I was least prepared for.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the requirements: latency, freshness, and consistency needs for the real-time feed. Then propose a dual-store architecture with a streaming ingestion pipeline, and explain how you ensure point-in-time correctness and low-latency serving.

Pro tip: Emphasize the trade-offs between freshness and cost/complexity, and mention how you would monitor feature drift and pipeline health to maintain freshness in production.

1. Clarify Requirements

Ask about latency SLAs, acceptable staleness, throughput, and consistency requirements for the real-time feed. This shapes the entire design.

2. Design Feature Store Architecture

Propose a dual-store setup: an offline store (e.g., data lake, warehouse) for training and batch features, and an online store (e.g., Redis, DynamoDB) for low-latency serving. Ensure feature consistency between stores.

3. Build Streaming Ingestion Pipeline

Use a stream processing framework (e.g., Kafka, Flink) to compute features in near real-time from event streams. Write to both online and offline stores, with backfill capabilities.

4. Ensure Training-Serving Consistency

Implement point-in-time correct joins for training data to avoid leakage. Use a feature registry to version features and share transformations between training and serving.

5. Monitor and Optimize

Set up monitoring for feature freshness, drift, and pipeline latency. Optimize by tuning batch sizes, caching, and using approximate algorithms where acceptable.

Key Points to Mention

  • Dual-store architecture (offline for training, online for serving)
  • Stream processing for real-time feature computation (e.g., Kafka, Flink)
  • Point-in-time correctness to prevent data leakage
  • Feature registry and versioning for consistency
  • Monitoring for freshness, drift, and pipeline health
  • Trade-offs between latency, cost, and complexity

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

Q5

How do you prevent feedback loops and ensure diversity in the recommendations over time?

Technical Trade-offsProduct Analytics & Metrics
Author's notes

Short answer from me: epsilon-greedy or UCB for exploration, and explicit diversity constraints at re-ranking.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging the feedback loop problem in recommender systems and its impact on diversity. Then, outline a multi-layered strategy combining algorithmic techniques (e.g., exploration, diversity constraints) with evaluation metrics and monitoring. Finally, emphasize the importance of continuous iteration and trade-off management between relevance and diversity.

Pro tip: Frame your answer around Atlassian's collaborative products: highlight how diversity in recommendations (e.g., suggesting varied Confluence pages or Jira issue types) can prevent echo chambers and foster cross-team knowledge sharing, aligning with Atlassian's values.

1. Define the problem and metrics

Explain how feedback loops occur (e.g., user interactions reinforce popular items) and define diversity metrics (e.g., intra-list similarity, coverage, entropy) to measure it.

2. Algorithmic interventions

Describe techniques like epsilon-greedy exploration, Thompson sampling, or diversity-aware re-ranking (e.g., MMR) to inject randomness and promote long-tail items.

3. Model and data strategies

Mention using debiasing techniques (e.g., inverse propensity scoring), incorporating content-based signals, and ensuring training data represents diverse user preferences.

4. Evaluation and monitoring

Set up A/B tests with diversity metrics, monitor for feedback loops (e.g., popularity bias), and use offline simulations to validate long-term effects.

5. Trade-off management

Discuss balancing relevance and diversity via multi-objective optimization or business rules, and iterate based on user feedback and business goals.

Key Points to Mention

  • Exploration vs. exploitation trade-off (e.g., epsilon-greedy, Thompson sampling)
  • Diversity metrics: intra-list similarity, coverage, novelty, serendipity
  • Debiasing techniques: inverse propensity scoring, position-based models
  • Re-ranking algorithms: Maximal Marginal Relevance (MMR), Determinantal Point Processes (DPP)
  • A/B testing and long-term holdout groups to measure diversity impact
  • Atlassian context: cross-team collaboration, avoiding echo chambers in Confluence/Jira recommendations

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