Data engineered into decisions.

Liresa builds governed data platforms that turn scattered operational signals into trusted, decision-ready stories.

Follow the story CV available on request

The data story

From noise to narrative.

01 Source work starts at the contract
Veeva, Oracle, SFTP, Fabric, survey, document, and media feeds get owners, cadence, schema expectations, and failure behavior before they land.
02 Governance is built into the lakehouse
Delta Lake and Unity Catalog carry permissions, lineage, validation results, and audit context through Bronze, Silver, and Gold layers.
03 Pipelines make quality visible
PySpark transformations, automated checks, profiling, and recovery paths turn raw movement into explainable data products teams can inspect.
04 Delivery stays close to users
APIs, Databricks Apps, dashboards, and migration tooling package the platform into workflows people can actually use.

Working proof

Every source has an owner, cadence, schema expectation, and failure path. Intake discipline

Proof

to 60% faster source onboarding through configuration-driven ingestion.

better data quality through automated profiling, validation, and lineage.

less manual analysis through NLP and extraction workflows.

Selected work

Databricks, Delta Lake, Unity Catalog, Workflows, Apps, ADF, and Snowflake.

Claude Sonnet, human approval, React, FastAPI, Databricks Apps, and Delta audit logs.

ETL, PostgreSQL, MongoDB, FastAPI, survey data, documents, NLP, and extraction flows.

Python, OpenCV, NumPy, Seaborn, Pandas, and research-driven defect detection.

Expertise

Field notes

Read the blog

Contact

Email LinkedIn