Staff Data Engineer, Core Migrations

Machinify is a healthcare intelligence company delivering AI-powered solutions that improve financial outcomes and reduce healthcare costs for health plans. The Staff Data Engineer on the Migrations team owns client migrations end to end, including legacy-system analysis, architecture, pipeline development, validation, deployment, stakeholder coordination, and post-launch handoff.

Responsibilities:

• Lead discovery & technical due diligence — get to the bottom of poorly documented legacy systems (ETL, stored procedures, reporting layers, file feeds), reconstruct the business logic, and capture it in the lineage maps, mapping specs, and risk analyses everyone builds from
• Reverse-engineer complex legacy systems using agents— you will be responsible for reverse-engineering complex, undocumented legacy systems (SSIS packages, stored procedures) where business logic is embedded, not documented. You must be able to reconstruct the intent of these systems to build modern, stable equivalents
• Drive ambiguity to resolution — spot the unknowns early, own the open questions, and pull answers from clients, SMEs, and Operations instead of waiting to be unblocked
• Architect & build the migration pipelines — turn intricate legacy logic into production-grade Airflow DAGs and Spark jobs on the Machinify platform, edge cases, payer-specific carve-outs, and business-rule exceptions included, owning the full flow from ingestion through reconciliation to steady-state handoff
• Make and document the hard architectural calls — own the decisions that matter (pipeline design, partitioning strategy, validation approach) and leave a clear trail of the reasoning so others can learn from and build on it
• Prove correctness at scale — build automated reconciliation frameworks that confirm, with confidence, that migrated output matches the source down to the row
• Own the program end to end — be the single technical owner from kickoff through go-live and hypercare: scope, sequence, and track the work, surface risks before they bite, align Operations, Client Success, Platform Engineering, SMEs, and the client, and drive UAT through to sign-off
• Raise the bar for the practice — Institutionalize migration knowledge by codifying runbooks and retrospectives, mentor L3/L4 engineers to build independent capability, and turn recurring migration pain points into scalable, reusable tooling

Requirements:

• 8+ years as a hands-on Data Engineer or Software Engineer, with demonstrated experience independently owning complex, multi-stakeholder technical projects from start to finish
• Strong Python and SQL — fluent with complex, unfamiliar legacy code, not just greenfield work
• Apache Spark — deep understanding of distributed processing, performance tuning, partitioning, and debugging at scale
• Apache Airflow — advanced; comfortable designing and authoring production DAG architectures from scratch
• AI-assisted development — actively uses coding agents (Claude, Copilot, Cursor, or equivalent) as a force multiplier in daily engineering work. Comfortable prompt-engineering for code generation, debugging, and documentation, and able to evaluate and trust AI output critically
• Legacy ETL stack — able to reconstruct intent from code and translate it faithfully into a modern stack (e.g., From SSIS, SQL Server stored procedures, T-SQL, or equivalent)
• Data validation & reconciliation at scale — experience designing automated frameworks that prove migrated data matches source systems with high confidence
• AWS — proficient with S3, solid grasp of cloud-native data patterns such as partitioned object storage, secure cross-account access, and cost/performance trade-offs
• Cross-functional coordination — proven ability to drive alignment across engineering, operations, and client-facing teams simultaneously, without losing velocity
• Clear technical communication — able to write discovery specs, architecture docs, and stakeholder updates that are accurate, concise, and actionable for both technical and non-technical audiences
• Data-first mindset & fidelity — a rigorous, data-first mindset, with deep experience in building automated reconciliation frameworks that guarantee 100% data fidelity between legacy and modern systems