Ocrolus is a document automation platform that powers the digital lending ecosystem, automating credit decisions across fintech, mortgage, and banking.
Some of my projects include:
Tabular Data Model and Tabular Data Flow (Java, Python, SQLAlchemy ORM, Postgres)
•Co-formed a 5 person cross-functional team with the mission of creating a new data model, service, and workflow that would allow for the decrease in TAT, handling of tabular data, deprecation of tech debt, and cannibalization of existing workflows
•Used Kubernetes to create, maintain and troubleshoot network issues regarding our new and existing services
Automation (Java, Python, Kafka, AWS Simple Queue Service, AWS Textract)
•Automated the process of identifying and populating data extracted from documents, decreasing the average time spent on a document by 50%. Created a roadmap for feature improvement and full automation, upon handoff of the project.
•Took ownership and improved legacy Form Parser Gateway (unified the output of various internal and external OCR parsers). Added parallelization, heuristics, muli-parser capabilities, standardized API contracts, and enabled tabular data handling
Validations (Python, TypeScript)
•Created a dynamic, REST-ful, and templatized system to validate the data we collect from documents. Backward compatible with legacy services. Allowed QA of automated and human tasks, informed heuristics to improve and speed up doc processing
Modification Logs (Python, Kafka, Snowflake)
•Created a generic and encapsulated logging system to track all steps in the identifying, capturing, and updating of data in the Tabular Data Workflow
•Worked with multiple departments to understand use cases and data requirements to build a single log table to capture all human and computer change-related data in our workflow
Worked closely with product to lead Agile ceremonies (scrum, backlog refinement, sprint planning)
