AI Production System: Data-Science Platform Modernization

The digital mortgage lender — the consumer mortgage business inside the parent company — is the U.S. online-mortgage leader, with trillions in originations since inception and recent acquisitions in real estate search and mortgage servicing. The technology platform is lender-built and lender-operated, AWS-heavy: SageMaker AI/Studio, EMR, S3, Athena, Lake Formation, Glue, Redshift, Lambda, EKS plus Istio, MLflow, model registry, feature store. Taller's engagement at the digital mortgage lender is the newest in the portfolio: active 2025–2026, sourced primarily from Argentina and Brazil, and anchored on individual contributor roles across Frontend, Backend, Cloud Infrastructure, Salesforce, and legacy modernization. Three cases below describe what Taller's engineering capacity supports at the lender.

The lender's data-science platform modernization moves the firm from a legacy on-prem Hadoop and vendor stack to managed AWS — SageMaker AI/Studio, EMR, S3, Athena, Lake Formation, Glue, Redshift, Lambda, EKS plus Istio for online scoring, MLflow, SageMaker Experiments, model registry, feature store. A multi-account VPC strategy isolates production with read-only access for build environments. The modernization requires substantial Cloud Infrastructure engineering capacity to operate at the scale the lender's data-science workload demands.

Taller's same-client engagement at the lender includes Cloud Infrastructure Engineering at the Lead and Senior levels, heavily weighted toward AWS — the dominant tech-stack signal across the lender roles. The capability reference Taller carries on MLOps platforms in regulated industrial environments is the manufacturing company engagement: AI Engineering, MLOps, and LLMOps work on a state-machine settlement workflow built on Java/Spring, the Temporal Framework, MySQL, REST endpoints, and GitHub Actions CI/CD, with an integrated risk-evaluation system.

Sustained Cloud Infrastructure engineering capacity at the lender on the AWS substrate the data-science platform requires.

The lender's data-science platform outcomes — delivery time for new data-ingestion projects collapsed from eight weeks to under one hour; a ninety-nine percent reduction in incident tickets over eighteen months; ten million automated decisions per day; five times growth in production models; eighty percent productivity gain for data scientists; model count from twenty-six to over two hundred ten — are the lender's outcomes earned by the lender's data-science platform engineering, MLflow / SageMaker / feature-store architecture decisions, and the in-house team operating the substrate. Capability fit between Taller's manufacturing company MLOps work and the lender's modernization is real; the platform substrate and the operating outcomes belong to the lender. Taller's contribution is the AWS Cloud Infrastructure engineering capacity supporting the workload.