OK
Ozair Khan
Senior Data Engineer • UIUC MCS Candidate

Architecting Enterprise Data Platforms & Applied AI Workflows.

Bridging production-scale distributed data engineering (BigQuery, Spark, Airflow) with modern Information Retrieval, RAG, and Agentic AI workflows.

Enterprise DW Migration High Impact
Oracle → BigQuery
Cloud Migration Lead architecting scalable ELT, partitioning & slot optimization.
Unified Data Product Production
Cost of Deal Platform
Consolidated multi-stream commercial, pricing, and ticketing telemetry.
Applied AI & Automation Bleeding Edge
DuckDB + Polars + MCP
Engineered local LLM metadata pipelines and diagnostic agentic workflows.
Section 01 // Production Track Record

Enterprise Architecture Case Studies

Deep-dive architectural RFCs detailing zero-downtime cloud migrations, multi-stream data pipelines, and verified operational benchmarks.

View All Case Studies →
Fintech & Commercial Intelligence Senior Data & Platform Engineer
Enterprise Verified

The 'Cost of Deal' Platform: Unified Enterprise Data Product

Consolidated fragmented pricing, ticketing, and merchant rebate pipelines into an immutable, idempotent Data Product serving C-suite margin analytics.

4.2x Faster
Financial Reconciliation Cycle
0.0%
Cross-Department Metric Drift
18M+
Daily Transactions Processed
99.4%
Data Quality Test Coverage
Apache Spark BigQuery Apache Airflow Python +3
Read Breakdown →
Enterprise Data Warehousing Lead Cloud Data Architect
Enterprise Verified

Enterprise Oracle to BigQuery Migration: Architecture & Cutover

Re-architected monolithic Oracle nightly batches into partitioned BigQuery ELT with Airflow DAGs, reducing critical reporting latency from 8 hours to 15 minutes.

-64%
Query Latency Reduction
$180K
Annual Cloud Compute Savings
2.4 TB
Daily Data Ingested
15 Mins
Data Freshness SLA
Google BigQuery Apache Airflow Oracle Python +3
Read Breakdown →
Section 02 // Engineering Labs & Research

Applied AI & Open-Source Tooling

Combining distributed data engineering discipline with UIUC CS 410/441 research, Model Context Protocol (MCP), and local LLM automations.

Explore All Labs →
Agentic Systems & MCP ● Active Lab

Building an MCP Server for Data Pipeline Triage & BigQuery Optimization

Empowering AI coding assistants with Model Context Protocol (MCP) tools to autonomously inspect BigQuery query execution plans, parse Airflow failure logs, and suggest partitioning and clustering fixes.

Python Model Context Protocol (MCP) Google BigQuery API Apache Airflow FastMCP AsyncIO
Media & Data Automation ● Completed

Automated YouTube & Media Asset Pipeline: FastAPI, DuckDB, Polars & Local LLMs

A production-grade local automation pipeline transforming raw video transcripts, timestamps, and slide benchmarks into SEO-optimized YouTube chapters, metadata, and cross-platform markdown notes.

Python FastAPI DuckDB Polars Ollama Whisper Pydantic
Information Retrieval & RAG ● Completed

Hybrid Search & BM25 + Dense Retrieval for Enterprise Data Catalogs

Why naive vector search fails on technical schemas and error logs, and how combining Okapi BM25 lexical ranking with dense semantic vectors (RRF) builds bulletproof RAG for Data Engineers.

Python UIUC CS 410 BM25 Vector Search Qdrant Cross-Encoders
1-on-1 Mentorship & Advisory

Transitioning into Data Engineering or scaling into AI Data Platforms?

I offer focused 1-on-1 advisory sessions for junior and mid-level engineers: architectural system design audits, cloud warehouse roadmaps (BigQuery/Spark), and mock technical interview preparation.