Lead Data Engineer + AI
Client - Altimetrik Takeda
Location: Remote
Need minimum 3 years of experience as Lead.
About the role
We're looking for a Senior Data Engineer to build and scale our Lakehouse and AI data pipelines on Databricks. You'll design robust ETL/ELT, enable feature engineering for ML/LLM use cases, and drive best practices for reliability, performance, and cost.
What you'll do
• Design, build, and maintain batch/streaming pipelines in Python + PySpark on Databricks (Delta Lake, Autoloader, Structured Streaming).
• Implement data models (Bronze/Silver/Gold), optimize with partitioning, Z-ORDER, and indexing, and manage reliability (DLT/Jobs, monitoring, alerting).
• Enable ML/AI: feature engineering, MLflow experiment tracking, model registries, and model/feature serving; support RAG pipelines (embeddings, vector stores).
• Establish data quality checks (e.g., Great Expectations), lineage, and governance (Unity Catalog, RBAC).
• Collaborate with Data Science/ML and Product to productionize models and AI workflows; champion CI/CD and IaC.
• Troubleshoot performance and cost issues; mentor engineers and set coding standards.
Must-have qualifications
• 10+ years in data engineering with a track record of production pipelines.
• Expert in Python and PySpark (UDFs, Window functions, Spark SQL, Catalyst basics).
• Deep hands-on Databricks: Delta Lake, Jobs/Workflows, Structured Streaming, SQL Warehouses; practical tuning and cost optimization.
• Strong SQL and data modeling (dimensional, medallion, CDC).
• ML/AI enablement experience: MLflow, feature stores, model deployment/monitoring; familiarity with LLM workflows (embeddings, vectorization, prompt/response logging).
• Cloud proficiency on AWS/Azure/GCP (object storage, IAM, networking).
• CI/CD (GitHub/GitLab/Azure DevOps), testing (pytest), and observability (logs/metrics).
Nice to have
• Databricks Delta Live Tables, Unity Catalog automation, Model Serving.
• Orchestration (Airflow/Databricks Workflows), messaging (Kafka/Kinesis/Event Hubs).
• Data quality & lineage tools (Great Expectations, OpenLineage).
• Vector DBs (FAISS, pgvector, Pinecone), RAG frameworks (LangChain/LlamaIndex).
• IaC (Terraform), security/compliance (PII handling, data masking).
• Experience interfacing with BI tools (Power BI, Tableau, Databricks SQL).