Omotosho Ayomide Quadri

Developed a PDF fraud detection system using Python (metadata tracker, NLP checks, etc), reducing fraudulent document submissions by 40%, and did a good number of integration tests (Pytest) to ensure 95%+ detection accuracy.

Collaborated with a team of 4 engineers to build a rule-based scoring engine that generates Thrift Scores by tracking user transaction patterns, spending behavior, and activity, improving risk assessment efficiency by 35%.

Engineered an automated amount generation algorithm tied to Thrift Scores, enabling dynamic group saving limits and increasing approval fairness by 25%.

Cleaned and processed 18+ commercial bank statements, using the data to extract key insights (income, cash flow, spending patterns) to support behavioral analysis and reduce manual review time by 60% using an ML engine.

Contributed to Circle AI, an LLM-powered tool for bank statement Q&A, improving customer support response time by 50% and reducing manual queries by 30%.

Built an analytics dashboard on Metabase for tracking revenue, user activity, marketing sources, and compliance, enabling data-driven decisions that boosted user retention by 15%.

Led the development of LENS 1.0 using a microservices architecture, with services handling business logic, loan applications, and a core loan engine for basic functionality.

Built CI/CD pipelines to automate deployment and integration processes, reducing manual errors by 25%.Configured a local Docker environment to run all microservices and set up staging-to-deployment workflows.

Developed several SQL pipelines to extract critical customer attributes (credit score, repayment history, income verification) from relational databases, accelerating risk assessment by 30% and improving ML model accuracy by 15%.

Implemented automated data validation checks in LENS (using Custom Python scripts) to verify data quality, consistency, and completeness before ingestion into pipelines, reducing errors to a large extent.

Integrated a payment gateway to process transactions for each loan recommendation.

Built and optimized a cloud-based data warehouse on AWS to centralize and securely store all loan application data, customer profiles, and transaction history for analytics and reporting.

Deployed Infrastructure as Code (IaC) using Terraform to automate the setup of proactive marketing features within the app, enabling dynamic customer targeting and personalized loan offers.

Automated data ingestion and transformation workflows using Azure Data Factory, enhancing pipeline reliability and reducing manual intervention.

Migrated large datasets from on-premises systems to Azure, ensuring 100% data integrity and minimal downtime.

Configured real-time streaming pipelines with Apache Kafka, enabling insights from live data streams for faster response times.

Designed and maintained Azure Synapse Analytics solutions to support enterprise data warehousing needs.

Conducted performance tuning for SQL queries, improving reporting speed by 40%.