~/neiman-analytics
I’m Phillip Neiman. I design and build the full analytics stack — ingestion, modeling, API, and interface — so teams stop arguing about numbers and start acting on them.
Built with
0+
Years in analytics
0+
dbt models in production
0.0%
Pipeline uptime
Selected work
Analytics intelligence for cosmetic dermatology practices. Tracks provider performance, channel attribution, patient recall, and no-show risk — from raw practice-management data through to a dashboard clinicians actually open.
The shared Python foundation underneath the products — config, connections, and pipeline primitives factored out so a new data product starts at week three instead of week one.
Capabilities
Warehouse design, ELT orchestration, and the contracts between them. Built to survive schema drift and staff turnover.
Dimensional models and metric layers that give one answer to one question — tested, documented, and versioned.
The last mile most data teams skip: APIs and interfaces that put the model in front of the person making the call.
LLM features grounded in your own warehouse — natural-language exploration and narrative summaries over governed data.
Most analytics work dies in the gap between a correct number and a decision someone actually makes. I build across that gap — warehouse through interface — because the handoffs are where the value leaks out.
Right now that means DermIQ, a vertical analytics product for cosmetic dermatology practices, and the shared platform underneath it.
Tell me what decision you’re trying to make. I’ll tell you what it would take to support it.