Data Engineer, Capgemini · Nashville, TN
Data and AI pipelines that fail loudly and cost little.
I’m Andrew Vernon. I replace manual, monthly processes with tested pipelines on Databricks, BigQuery and AWS, and I build LLM extraction you can audit.
Selected work
Two side projects, each built under a hard $15 monthly cap. Both are in development.
Onchain Indexer
In developmentAn Ethereum event indexer that keeps up with the finalized head on a $15 a month AWS budget.
- Python
- Postgres
- AWS
- Terraform
- pytest
EDGAR Events
In developmentSEC 8-K filings turned into typed, entity-resolved event rows, with evals and a cost ledger you can audit.
- Python
- Claude API
- AWS Step Functions
- Lambda
- DuckDB
I look for the process someone runs by hand every month, then replace it with a pipeline that is tested, cheap to run and boring to operate.
- Based
- Nashville, TN
- Now
- Data Engineer, Capgemini
- Certified
- Databricks Data Engineer Professional, Google Cloud Professional Data Engineer
Data, AI and software
Three disciplines with one job: reliable answers out of messy sources.
01 · Data
Pipelines and modeling
Batch and streaming ingestion, medallion layers, and contracts that stop bad rows before they spread.
- Python
- SQL
- PySpark
- Databricks
- BigQuery
- SQL Server
- Postgres
- DuckDB
02 · AI
LLM systems you can measure
Extraction and entity resolution backed by golden sets, judges and a cost ledger, so accuracy and spend are both on the page.
- Claude API
- Evals
- Entity resolution
- Embeddings
- Versioned enrichment
03 · Software
Shipping and operations
Tests, CI, infrastructure as code and alerting. Backups get restore drills and schemas get diffed against the repository.
- pytest
- GitHub Actions
- Terraform
- AWS
- Google Cloud
- Azure
- Linux
A $15 monthly budget teaches more engineering than an unlimited one.
Results at work
Automated $75k of annual labor with a pipeline that reads project updates out of unstructured email and writes them to a project database.
Removed 80% of the noise from a network access-failure metric by working through its definition with regional engineering teams.
Took reporting from weekly to near real-time by unifying human-labeled data and automated extraction in a medallion analytics database.
Automated 5 legacy processes as owner of architecture and modeling for two core pipelines landing BigQuery data in SQL Server.
About
I’m a data engineer in Nashville. Since 2022 I’ve built the data behind decisions at T-Mobile, JLL and now Capgemini, where I lead architecture and modeling for a national nonprofit’s data platform.
Outside work I build under tight constraints on purpose. A hard cost cap forces real trade-offs, and it leaves a cost line anyone can check.
- Databricks Certified Data Engineer ProfessionalAug 2026
- Google Cloud Professional Data EngineerJun 2026