Data Engineer | Data Warehouse Specialist | Builder of Reliable Data Systems
I design and build modern data warehouses and analytics platforms that actually make sense — the kind that analysts enjoy querying and engineers enjoy maintaining.
Most of my work lives somewhere between data modeling, pipeline automation, and performance tuning.
- Architect and scale data warehouses for analytics and reporting
- Build ETL/ELT pipelines using DBT, PySpark, and AWS Glue
- Design dimensional models and data marts that power self-serve dashboards
- Automate metadata, lineage, and documentation (because we all forget to update README files)
- Tune Redshift clusters until they behave like they should’ve from day one
| Area | Tools & Tech |
|---|---|
| Data Warehouse | Redshift, Snowflake, Postgres |
| Transformations | DBT, PySpark, Glue |
| Data Modeling | Kimball, Data Vault, Star Schema |
| Metadata & Governance | OpenMetadata, Amundsen |
| Infrastructure | AWS, Docker, Terraform, Kubernetes |
| Languages | Python, SQL, a bit of TypeScript |
- Central Data Warehouse: Unified Finance, Ops, and Product data into one source of truth; reduced report latency by 80%.
- Event Pipeline: Designed a Kinesis → Lambda → Redshift pipeline handling 50M+ daily events.
- Metadata Automation: Integrated DBT with OpenMetadata for end-to-end lineage and documentation.
- Metadata-driven warehouse automation
- Using LLMs to generate data documentation and quality tests
- Lightweight internal dashboards using Svelte and FastAPI
- “SELECT *” is fine — as long as you know why you’re doing it.
- A well-modeled schema beats any fancy dashboard.
- The best pipelines are the ones you forget exist because they never break.
- Portfolio: krishnanandanil.com
- LinkedIn: linkedin.com/in/krishnanand-anil
- Email: krishnanandpanil@gmail.com
“Good data models are like good jokes — if you have to explain them, they’re not working.”
If you see something interesting here, clone it, break it, and make it better. That’s how most of my projects started anyway.