Senior Data Engineer

We're hiring an AWS data engineer to lead the hands-on build on a large data migration: moving data out of legacy systems into a modern lakehouse on AWS, without breaking the reporting the business runs on while it happens.

Migration work is often treated as grunt work. We don't. Done properly it's some of the most demanding engineering in data — reconciling systems that were never meant to agree, proving to the byte that nothing was lost, and designing the target so the client is better off, not just relocated. That's the standard you'd be working to.

What you will do

• Design and build migration pipelines in PySpark and Python on AWS — Glue, EMR, S3, Redshift, Lambda and Step Functions
• Profile legacy sources, map schemas to the target model, and handle the mess that mapping always uncovers
• Write reconciliation and validation checks that prove source and target match, and make the evidence reviewable
• Plan and run cutovers with rollback paths, so go-live is an event, not a gamble
• Tune Spark jobs for large volumes — partitioning, memory, cost — because migration windows are finite
• Work directly with the client's engineers and stakeholders; there is no layer of managers between you and them
• Document as you build: runbooks, mapping specs and decisions, written so the client can own the result

What we look for

• Strong PySpark, Python and SQL — you can reason about a query plan and a Spark DAG, not just write code that runs
• Production experience on AWS with real data volumes; certification is nice, shipped work matters more
• Someone who has lived through at least one migration, upgrade or replatforming and remembers what went wrong
• Care for data correctness bordering on stubbornness — reconciliation is the job, not an afterthought
• Clear written English; half our communication with clients is written

Nice to have

• Glue Data Catalog, Lake Formation, Athena or Iceberg/Delta table formats
• dbt for the post-migration transformation layer
• CDC tooling such as DMS or Debezium
• Terraform or CloudFormation

Why this role, honestly

• We are four senior people; you would be the fifth. Every voice shapes how we work
• You own your work end to end — the person who designs the pipeline runs the client demo and answers for it in production
• Certifications are paid for, with study time given, not squeezed into weekends
• No internal politics to navigate, because there is no internal
• The flip side, stated plainly: small company, client work, real deadlines. If you want a big-company cocoon, this isn't it

How we hire

• Short intro call, then one technical conversation about a migration scenario — no leetcode, no take-home longer than an hour
• You'll meet all four of us before any offer
• Every application gets a reply, including the ones we turn down