Applied ML engineer building APIs, batch workflows, and practical data systems.

If you need dashboards, reports, or spreadsheet automation, see Services.

If you need ML systems, APIs, or batch workflows, start with Portfolio.

Proof of work

Public repos and live demos from the portfolio and data tools tracks.

Customer churn prediction system
  • FastAPI
  • scikit-learn
  • Classification
  • Joblib

Serves per-account churn probability, tier, and routing flag from one pinned model and cutoff.

RAG document intelligence QA
  • FastAPI
  • FAISS
  • sentence-transformers
  • Docker

Ingest chunks documents for retrieval, then answers questions using retrieved context and a hosted model.

Data cleaning toolkit
  • Streamlit
  • pandas
  • pytest
  • Parquet / JSON

Upstream prerequisite, not a side utility: downstream models and dashboards ingest the same reviewed tables—multi-format inputs become auditable CSV/Parquet/JSON plus HTML step log and before/after views, capped near 100K rows; rules cover bad formats, duplicates, skewed categories, optional outliers, plus bundled samples for dry runs.

KPI dashboard app
  • Streamlit
  • Sales & marketing CSV
  • Mapping presets
  • Optional OpenAI

Metric tracking and stability—not narrative exploration: preset mappings and hard row gates drive KPI cards, trends, breakdowns, optional deltas, and a rule-first “what changed?”; optional OpenAI reads only pre-aggregated KPI objects, never raw CSV rows.

Applied ML, APIs, and practical data systems.

Scoped work for ML systems, dashboards, and repeatable reporting. Full offers and detail are on the Services page.

Dashboard & KPI reporting
KPIs, trends, and breakdowns in Excel, Google Sheets, or Power BI—a single view built on your data, with light handover your team can maintain.

Typical project range: €300 – €900

DashboardKPI definitionsdocumentationExcel
Reporting & spreadsheet automation
Stop rebuilding the same report every week. Scheduled or on-demand pipelines—in Sheets, Apps Script, or Python—that mirror the discipline behind batch scoring: same inputs, same rules, reproducible outputs.

Typical project range: €200 – €700

Automated workflowsscriptsdocumentationExcel

Final pricing depends on data complexity and project scope.

How I Work

1. Define scope and outputs
We agree what to solve, what goes in, what must come out, and whether the core is an API, batch work, or a smaller reporting slice. Scope is clear in writing before anything is built.
2. Review data and limits
I check data shape, quality, and hard limits—what is reliable, what is missing, and what is out of scope. You get a straight answer on what the data can support.
3. Build the system
I build reproducible pipelines or services with clear steps. APIs and batch jobs come first when they fit; dashboards or extra reporting only when they belong in that system.
4. Validate and deliver
I test what matters for the agreed scope, keep runs repeatable under the same rules, and leave short notes your team can follow. You receive something you can run and own—not a throwaway demo.

Core stack

PythonSQLAPIsFastAPIBatch processingDockerpytestscikit-learnExcel, Sheets, Power BI, Looker Studio

Scope and fit

Send goal, current stack, data sources, and timeline. I will say whether the work matches applied ML, APIs, and batch workflows, or a narrower reporting and data-prep scope. Use the contact form to send this.

© 2026 Vahdettin Karataş. All rights reserved.
Applied ML systems, APIs, and practical automation.