Customer Success Analyst

Point Predictive powers fraud prevention for auto and consumer lenders through the industry’s largest lending fraud data consortium. When one lender sees a fraud pattern, every lender in the consortium benefits — a “herd immunity” model that stops synthetic identities, straw borrowers, income and employment misrepresentation, and dealer-facilitated fraud before they become losses. Job Summary: This is a customer-facing analyst role. The analysis you build exists to serve customers, and you will present findings, troubleshooting their data feeds, and help turn what the data says into action. We're hiring for someone equally strong in both halves of the job: the technical depth to build it, including real fluency with AI and automation, and the people skills to deliver it. The Customer Success Analyst is responsible for ensuring our fraud detection models work correctly for every customer. Keeping models performing at their best requires constant vigilance – watching for model drift, distribution shifts, and changes in input data quality. This role owns monitoring those signals and works with customers or the data science team to get issues resolved. This role also owns company reporting: building and automating the data behind everything from board-level internal reporting, to fraud strategy analysis, to external monthly reports sent to customers. This is an in-person role based at our San Diego headquarters. Core Responsibilities: Customer Partnership & Communication (50%) Monitor account data health proactively. Know which customers' data is healthy, which is slipping, and why, before it becomes a customer-facing problem Work directly with customers to problem-solve when their input data is missing, incomplete, or inaccurate. You own the issue from detection through resolution. Prepare and present data and analysis for quarterly business reviews, monthly business reviews, and executive/board reporting packages Translate complex data findings for customers, executives, and engineers alike. Partner with Customer Success Managers, fraud strategists, data science, and engineering to get the right people on each customer issue and keep everyone aligned to one outcome Technical Ownership (50%) Build and monitor customer data quality, score distributions, fill rates, alert firing rates, and model performance (PSI, AUC, score drift) using Sigma, Validio, and other AI tools Automate relentlessly. Replace manual reporting and monitoring processes with scheduled pipelines and AI-assisted workflows so coverage scales without headcount Design, build, automate, and maintain internal and external-facing reports and dashboards Analyze fraud and default data to help build loan-automation and fraud-prevention strategies for customers What You Bring Experience in data analytics, analytics engineering, or a related role Experience in a customer facing role Fluency with AI tools as part of your daily workflow: using LLMs to speed up analysis, build monitoring agents, and automate work that used to be manual Proficiency in Python for data pipelines, automation, and reporting Experience with BI tools (Sigma, Tableau, Power BI, etc) Detail-oriented with a keen eye for spotting anomalies Ability to present complex data findings to both technical and non-technical audiences Ability to work independently, own project outcomes, and relentlessly problem-solve A customer-first mindset: comfort picking up the phone, running a meeting, and being the person a customer trusts with their data You live our core values: be the expert, pitch in, and get it done. Preferred Qualifications BA or BS in Computer Science, Data Science, or a related field Familiarity with machine learning concepts and how model performance is measured in production Experience with dbt or comparable data transformation frameworks (dbt Fundamentals course completion) Experience in lending, fraud detection, or financial services data Experience working directly with customers or external stakeholders on data issues Familiarity with AWS services, particularly CloudWatch and log analysis #J-18808-Ljbffr Point Predictive