Analytics
DataPulse
An end-to-end Streamlit analytics dashboard spanning ingestion, transformation, business KPIs, visualization, export, and a forecasting path.
- systems programming
- data
- databases
- product
- people
01
The product shape
I treat DataPulse as evidence of an analytics workflow more than as an ML claim: configured ingestion, transformation, KPIs, charts, export, targets, and forecasting connected in one application.
02
What I can show
I have concrete repository support for Pandas-based transformation, configurable column mappings, service boundaries, visual reporting, and a small known-data KPI test.
03
What I do not claim
I have not verified forecast evaluation, model comparison, a messy external dataset study, or production usage. I therefore call this analytics with forecasting—not AI research.
04
Next proof
I would strengthen this as a data-science case study with a reproducible public-safe dataset, runtime validation, baseline comparison, and explicit error metrics.
Technical notes
Core stack
Python
Pandas
Streamlit
Plotly- Prophet
Deployment & services
Streamlit Community Cloud
Supabase
I deployed this through Streamlit Community Cloud with Supabase-backed services.
View public repository