Full Stack Engineer Challenge
Reconciling Financial Transactions - Design a small, reliable transaction reconciliation feature
Create a free account to upload your work. Your progress saves as a draft until you submit.
What You'll Be Doing
Scenario — "Reconciling Transactions Across Accounts"
A company's platform ingests financial transaction data from multiple sources — banks, CRMs, and internal ledgers. Build the first version of a Transaction Reconciliation Service that helps finance teams identify mismatches between imported and expected transactions.
Design a simple full-stack feature that:
- Ingests transactions from two sources — e.g., bank feed and internal ledger
- Compares transactions using key fields (date, amount, merchant, reference)
- Flags mismatches or duplicates for review
- Displays results in a clean frontend view that supports sorting and filtering
- Supports auditing — every match or mismatch is logged for future reference
Submit a system design sketch, a UX and validation plan, and a video walking through your approach.
Challenge Guidelines
- Video: 7–10 minutes (continuous recording, one take)
- Written Artifacts: Two short (1 page max each) documents
- Scenario: Design a small, reliable transaction reconciliation feature
- AI Usage Declaration: Required
What You'll Accomplish
Design data models for financial reconciliation
Build reliable matching algorithms for transaction data
Handle edge cases like rounding errors and time zones
Create transparent, auditable UX for fintech applications
Balance performance with data integrity
How Your Work Will Be Scored
What to Submit
Required Submissions:
- System Design Sketch (Required)
- Format: PDF or Document
- Length: 1 page maximum
- Outline your architecture and data flow:
- How you'd model transactions and reconciliation results (PostgreSQL schema or ORM example)
- What backend endpoints you'd define (POST /reconcile, GET /results, etc.)
- How you'd ensure data integrity and handle partial matches or rounding errors
- Keep it high-level — just enough to see how you'd reason about correctness and scale
- UX & Validation Plan (Required)
- Format: PDF or Document
- Length: 1 page maximum
- Describe how you'd make this feature clear and trustworthy for end users:
- How would the UI communicate matches, mismatches, or pending reconciliations?
- What validation or error-handling logic would you add on the frontend?
- How would you track and log reconciliation history for audits?
- Focus on transparency and user confidence — key traits in fintech systems
- Video Recording (Required)
- Format: MP4
- Length: 7-10 minutes (continuous recording, one take)
- Must include all three components detailed below
Your video must include ALL three components:
1. Introduction (1-2 minutes)
- Share one example of a time you built or optimized a financial or data-heavy feature end to end. How did you balance performance, correctness, and maintainability?
2. Traits Assessment (2-3 minutes)
Which of these situations fits how you like to work — and why?
- Scenario A — "The Missing Match" (Execution): A customer reports that certain incoming transactions aren't matching expected records in the dashboard. You have clear repro steps, database access, and error logs. You need to debug and fix the reconciliation logic quickly without breaking anything else. Would this type of focused, high-ownership debugging energize you?
- Scenario B — "The Future Flow" (Discovery): You're asked to design a new, automated reconciliation process that scales as customers grow 10×. You'll research data models, design APIs, and test with synthetic data before release. Would you thrive in this open-ended, design-oriented work?
There's no "correct" choice — just share which environment helps you do your best work.
3. Challenge Response (3-5 minutes)
Walk us through how you would design and implement a minimal reconciliation feature:
- What data models or schemas would you use?
- How would the backend API work?
- How would you present the data in the frontend for clarity and trust?
AI Usage Declaration:
At the end of your video, please state:
- ☐ I did not use any AI tools
- ☐ I used AI tools (briefly describe how, including prompts and edits)
System Design Sketch
Format: .pdf, .doc, .docx, .rtf, .txt, .md
Outline your architecture and data flow:
- How you'd model transactions and reconciliation results (PostgreSQL schema or ORM example)
- What backend endpoints you'd define (POST /reconcile, GET /results, etc.)
- How you'd ensure data integrity and handle partial matches or rounding errors
Keep it high-level — just enough to see how you'd reason about correctness and scale.
Sign in to upload files
UX & Validation Plan
Format: .pdf, .doc, .docx, .rtf, .txt, .md
Describe how you'd make this feature clear and trustworthy for end users:
- How would the UI communicate matches, mismatches, or pending reconciliations?
- What validation or error-handling logic would you add on the frontend?
- How would you track and log reconciliation history for audits?
Focus on transparency and user confidence — key traits in fintech systems.
Sign in to upload files
Video Recording
Format: .mp4, .mov, .webm
Record a continuous, one-take video covering all three components:
1. Introduction (1-2 minutes)
Share one example of a time you built or optimized a financial or data-heavy feature end to end. How did you balance performance, correctness, and maintainability?
2. Traits Assessment (2-3 minutes)
Which of these situations fits how you like to work — and why?
- Scenario A — "The Missing Match" (Execution): A customer reports that certain incoming transactions aren't matching expected records in the dashboard. You have clear repro steps, database access, and error logs. You need to debug and fix the reconciliation logic quickly without breaking anything else. Would this type of focused, high-ownership debugging energize you?
- Scenario B — "The Future Flow" (Discovery): You're asked to design a new, automated reconciliation process that scales as customers grow 10×. You'll research data models, design APIs, and test with synthetic data before release. Would you thrive in this open-ended, design-oriented work?
There's no "correct" choice — just share which environment helps you do your best work.
3. Challenge Response (3-5 minutes)
Walk us through how you would design and implement a minimal reconciliation feature:
- What data models or schemas would you use?
- How would the backend API work?
- How would you present the data in the frontend for clarity and trust?
AI Usage Declaration
At the end of your video, please state one of the following:
- I did not use any AI tools
- I used AI tools (briefly describe how, including prompts and edits)
Sign in to upload files
Create a free account to upload your work. Your progress saves as a draft until you submit.
On this page