Abhishek Singh

Designed and built the end-to-end data platform powering enlytical.ai, an AI-enabled ads & analytics platform.

• Built a distributed data-collection platform orchestrating ingestion across 50+ heterogeneous APIs and web sources, landing raw data into a Delta lake for downstream processing.
• Built a centralized job-orchestration service that schedules, dispatches, and monitors crawling and data-processing workloads across 100+ Azure VMs.
• Built a distributed, event-driven ingestion platform on Azure Storage Queue events with asynchronous, fault-tolerant execution — cataloguing and processing terabyte-scale data through medallion layers.
• Implemented a Delta Lake medallion architecture on ADLS with file-level partitioning and scheduled optimization to cut query scans and storage costs.
• Built a config-driven ingestion framework that provisions schemas, transformations, validation rules, and storage layouts for new datasets without code changes — plus a metadata-driven data-reliability suite guarding every layer.
• Built a cross-cloud serving layer via Synapse/Fabric and BigQuery, exposing curated gold datasets to product dashboards and FastAPI services.
• Engineered a low-latency analytics layer with pre-aggregations, query-optimized storage, and in-memory caching — bringing interactive dashboard responses to sub-second.

Owned data architecture, SQL administration, and analytics development.

• Initiated the migration off legacy SQL Server, standing up the first lakehouse foundation — Parquet on ADLS with a Synapse Serverless SQL serving layer — that later grew into the production platform.
• Built and maintained complex SQL- and Spark-based ETL pipelines, including stored procedures and functions, to serve business-critical analytics dashboards.
• Orchestrated data workflows and refresh schedules while managing SQL Server performance, uptime, and processing capacity.