At Zest AI, the Business Analyst role is a highly strategic, high-impact position that sits at the intersection of data science, financial services, and client success. Zest AI is a leader in artificial intelligence-driven credit underwriting, helping credit unions and banks make faster, fairer, and more accurate lending decisions. As a Business Analyst, you are not just crunching numbers; you are translating complex machine learning models into tangible business value for financial institutions.

Your primary mission is to analyze loan portfolios, build predictive business cases, and demonstrate how transitioning from traditional credit scoring to Zest AI's machine learning models can safely increase approval rates while reducing default risk. This requires a unique blend of deep analytical rigor, financial acumen, and exceptional communication skills, as you will frequently present your findings directly to both internal product teams and external executive stakeholders.

This role is critical to the company's growth because it directly influences client adoption. By proving the financial and statistical validity of Zest AI's technology on real-world portfolios, you help drive the transition toward more equitable and efficient lending. It is an intellectually challenging and deeply rewarding position where your insights directly impact the financial lives of millions of borrowers.

The interview process at Zest AI evaluates your analytical capabilities, statistical knowledge, and communication skills. The following questions are representative of what you can expect, compiled from real candidate experiences. They are designed to test your structured problem-solving and ability to communicate complex concepts clearly.

Finance, Statistics, and Case Analysis

This category tests your ability to interpret credit risk metrics, work with financial data, and apply statistical concepts to business problems.

• How would you evaluate the performance of a credit scoring model if the historical data has a high rate of missing values?
• Given a dataset of loan applicants, how would you determine the optimal cut-off score to maximize profit while keeping defaults below a specific threshold?

To succeed in the Zest AI interview process, you must prepare to demonstrate a balance of technical execution and strategic communication. Interviewers want to see how you think, how you handle ambiguity, and how you present your ideas under pressure. Focus your preparation on the following core evaluation criteria.

Analytical & Statistical Rigor – You must be comfortable working with large datasets, understanding probability, and applying statistical concepts to credit risk. Expect to be tested on your ability to interpret data patterns and draw logically sound conclusions.

Structured Problem-Solving – When faced with complex, ambiguous business cases, you need a structured framework to break down the problem. Your interviewers will evaluate how logically you construct your hypotheses, clean your data, and build your financial models.

Client-Facing Communication – A great analyst at Zest AI is also an educator. You must be able to translate complex machine learning concepts and statistical outputs into clear, actionable business strategies that non-technical executives can easily understand.

Technical Execution – You will need strong hands-on skills in Excel, SQL, and potentially Python or R. You must be able to manipulate data quickly and accurately during self-administered case studies and take-home projects.

The interview process for a Business Analyst at Zest AI is highly comprehensive and realistic, designed to mirror the actual day-to-day responsibilities of the role. Rather than relying solely on abstract brainteasers, the company focuses heavily on practical homework assignments and case presentations.

The typical progression consists of four distinct stages:

• Initial Recruiter Screen: A standard introductory call to discuss your background, interest in Zest AI, and high-level behavioral alignment.
• Self-Administered Case Study: A timed, 90-minute technical assessment focused on finance and statistics, typically completed in Excel.
• Technical Case Interview: A deep-dive discussion with a senior analyst or the head of the Business Analyst team to review your case study methodology and problem-solving approach.
• Final Round Panel & Presentation: A comprehensive final stage where you present a take-home project (which may involve machine learning concepts) to a panel, followed by individual 1-on-1 interviews and behavioral evaluations.

The timeline above details the typical progression from the initial application to the final decision. Candidates should use this sequence to pace their preparation, ensuring they master Excel and basic credit risk concepts before the second stage, and refine their presentation skills ahead of the final round. While the exact duration can vary based on scheduling, the process is designed to be highly transparent and communicative throughout.

To stand out in the Zest AI hiring process, you must excel in three core evaluation areas that directly correspond to the stages of the interview loop.

Business Case Analysis & Modeling

This area evaluates your ability to handle raw data, apply statistical methods, and build reliable financial models. This is primarily tested during the 90-minute self-administered case study and the subsequent technical interview.

Be ready to go over:

• Excel Modeling: Building clean, dynamic models to forecast loan performance, approval rates, and default costs.
• Credit Risk Metrics: Understanding key lending concepts such as charge-offs, loss ratios, approval rates, and risk-adjusted return on capital (RAROC).
• Data Validation: Identifying anomalies, missing variables, or biases in a historical lending dataset.
• Advanced concepts (less common): Machine learning model evaluation metrics (e.g., Gini coefficient, KS statistic, AUC-ROC) and how they relate to traditional credit scoring.

Example scenarios:

• "Analyze this sample loan portfolio in Excel and determine the financial impact of lowering the credit score cutoff by 20 points."
• "Explain how you would handle a dataset where 30% of the applicants have no credit history (thin-file borrowers) when building a risk model."

Project Presentation & Client Communication

A critical component of the final round is presenting a take-home project to a panel of Zest AI team members. This stage simulates a client-facing delivery where you must defend your analytical choices and explain complex concepts clearly.

Be ready to go over:

• Structured Slide Design: Creating a concise, visually clear presentation deck that highlights key business takeaways first.
• Handling Rigorous Q&A: Answering challenging questions about your data assumptions, model inputs, and statistical methodologies without losing your composure.
• Translating Tech to Business: Explaining how machine learning algorithms (like gradient boosted trees) improve upon traditional logistic regression models in a way that a business executive can understand.

Example scenarios:

• "Present your findings from the take-home machine learning project, explaining why you chose specific features and how your model impacts the client's bottom line."
• "During your presentation, an interviewer playing the role of a skeptical client asks: 'Why should I trust your AI model over my 20 years of underwriting experience?' How do you respond?"

Behavioral & Collaboration Dynamics

Zest AI values highly collaborative individuals who can work cross-functionally with data scientists, software engineers, and account executives. They look for candidates who are intellectually curious, open to feedback, and highly professional.

Be ready to go over:

• Receiving Feedback: Demonstrating how you handle constructive criticism of your analytical models or presentation style.
• Cross-Functional Collaboration: Showing how you translate technical requirements between data scientists and business teams.
• Ambiguity Management: Explaining how you approach projects when the data is messy or the business objectives are not fully defined.

Example scenarios:

• "Tell me about a time when you realized your initial analysis was incorrect mid-way through a project. How did you pivot?"
• "Describe a situation where you had to explain a highly technical limitation to a client who wanted immediate results."