Senior Analytics Engineer (Business Intelligence & Product Analytics) · Snowflake · dbt · Looker · SQL · Python
I'm a Senior BI / Analytics Engineer with 6+ years of experience building analytics infrastructure that Finance, Product, and Marketing teams actually trust and use.
My focus is the layer between raw data and business decisions governed semantic layers, clean dbt models, and dashboards (and increasingly, AI interfaces) that answer the question before someone has to ask it.
| Layer | Tools |
|---|---|
| Data Warehouse | Snowflake |
| Transformation | dbt Core |
| Semantic Layer | dbt MetricFlow, Looker (LookML) |
| BI / Visualization | Looker, Tableau, SSRS |
| Languages | SQL, Python |
| Backend / AI | FastAPI, RAG (ChromaDB), LLM orchestration |
| Cloud | AWS |
| Other Databases | SQL Server, Oracle |
- Built SemanticGateway a governed semantic layer that translates natural language into validated Snowflake SQL via MetricFlow, with zero hallucinated joins or grain violations
- Modeled Customer LTV analytics across 1M+ customers — driving retention and segmentation decisions for Finance and Marketing
- Built 50+ executive and financial dashboards spanning revenue, margin, forecasting, and product performance
- Improved forecasting accuracy by 35% through analytical modeling and cleaner upstream data contracts
- Led self-service analytics adoption via semantic modeling, LookML governance, and stakeholder training
Live app: https://semanticgateway.vercel.app
A governed semantic layer that turns natural language into validated Snowflake SQL — without hallucinated joins, metric misuse, or grain violations. The LLM handles intent extraction and narrative generation only; all SQL is generated by MetricFlow from a governed semantic manifest, then reviewed for grain mismatches and fan-out risk before it ever touches Snowflake. Built on a synthetic streaming-SaaS dataset (1.8M+ rows, 18 certified metrics, 30 dbt models, 79/79 tests passing), with RAG-based metric retrieval, a three-tier LLM fallback chain, and a two-layer cache bringing warm queries down to ~1–3 seconds. Ships with an offline eval harness for scoring intent-extraction accuracy against a pinned golden set.
FastAPI MetricFlow dbt Snowflake RAG React OpenRouter
Live app: https://lookml-auditor-web.vercel.app
LookML Auditor is a powerful, privacy-first static analysis tool for Looker projects. It helps developers maintain high quality code by detecting broken references, duplicate definitions, and join integrity issues all in one beautiful dashboard.
Python looker parser lookml react
Analytics pipeline for a synthetic SaaS dataset — synthetic event data generated with Python, modeled in dbt, and structured to answer product and growth questions like activation rates, feature adoption, and churn signals.
Python dbt Product Analytics Churn Activation
End-to-end analytics engineering project on the Instacart dataset. Built on Snowflake + dbt + LookML with a full star schema, customer RFM segmentation, cohort retention analysis, and a semantic layer with reusable explores and metrics.
Snowflake dbt LookML Star Schema RFM Cohort Analysis
- Agentic analytics interfaces — extending SemanticGateway's pattern (governed semantic layer + LLM orchestration) to more complex, multi-step analytical workflows
- Python for analytics engineering — data profiling, pipeline scripting, and pandas for pre-modeling EDA
- Fraud & fintech data modeling — signal design and behavioral feature engineering for risk and AML use cases
- Google Business Intelligence Certificate
- Microsoft Certified: Azure Database Administrator Associate