Tech Lead - Data Platform interview questions

I've been in data engineering for nine years, three of them leading a data platform team in Oman. Started in ETL development at an Indian fintech, moved into modern data stack work around 2020, and for the past three years I've led the data platform team at an Omani Tier-1 bank: 12 engineers split between data engineering, analytics engineering, and platform infrastructure. We run a Databricks lakehouse on Azure serving 50+ downstream use cases. I report to the head of data and report monthly to the executive on data initiatives.

Medallion architecture: bronze (raw, immutable), silver (cleaned and conformed), gold (business-aggregated for consumption). Each layer with clear ownership and SLAs. Streaming and batch coexist; not everything needs to be streaming. Strict schema management; Delta Lake's schema enforcement prevents silent breakage. Self-service for analysts within the gold layer; tight access controls on bronze and silver. Cost management built in: tags, query monitoring, alerts on runaway compute. Documentation in the catalog, not in random wikis.

Quality gates at every layer. Bronze: source connection and freshness checks. Silver: schema validation, referential integrity, business rule checks (revenue can't be negative, dates must be valid). Gold: reconciliation against source-of-truth systems, trend checks (sudden 50% drop in a metric is suspicious). Tools like Great Expectations or Soda Core for explicit assertions. Failures block downstream usage; we don't serve bad data even if late beats wrong. Monthly data quality review with stakeholders; trust is earned, not declared.

Strict requirements. Data classification at ingestion: PII, financial, public. Access controls per classification using Unity Catalog with role-based grants. PII masking in non-production environments automatically. Audit logging on every PII access. Lineage tracking so we can trace any field back to source. Right-to-erasure designed in for GDPR-style requests. Regular access reviews with business sponsors confirming who needs what. Bank regulators audit data practices seriously; governance can't be retrofitted, it must be designed in.

Service mindset. The platform exists for analysts and scientists, not the other way around. I prioritise their unblocking, document datasets well, respond fast to data issues. I also push back constructively: ad-hoc queries that become recurring patterns get productised as proper datasets; analysts shouldn't have to write the same SQL repeatedly. Joint planning sessions quarterly so I understand what's coming and they understand my capacity constraints. Bad data platform teams treat downstream as customers to be tolerated; good ones treat them as partners.

For a tech lead role on a bank data platform in Oman I'd target OMR 2,500 to 3,200 total package depending on the team scope and platform complexity. Banking-context data leadership commands a premium. I'm on 90 days' notice. Beyond pay I'd value the data maturity of the organisation; data leadership in an org that doesn't trust data isn't satisfying regardless of pay.