Welcome to my GitHub 👋

I’m a Data Scientist focused on building reproducible, decision-support data products across data preparation, EDA, statistical analysis, machine learning, business intelligence, and GenAI-oriented workflows. My work centers on framing the right problem, defining meaningful metrics, and translating analysis into clear, actionable business outcomes.

My background combines quantitative rigor from Mathematics & Science Education, human-centered analysis from Sociology, and systems + business thinking from Management Information Systems (MIS) with MBA-level coursework. I work with a documentation-first, reliability-focused mindset shaped by PMI-aligned project discipline, structured analysis, and stakeholder-oriented communication.

All projects, demos, and technical write-ups live here:
➡️ github.com/totkanligizem/portfolio

I treat each project as a mini product:
clear objective → measurable metric → reproducible workflow → delivery-ready artifact

• Problem framing, hypothesis design, and dataset strategy
• EDA, feature engineering, modeling, evaluation, and error analysis
• Leakage-safe validation with train-only fit discipline and CV hygiene
• Reproducible workflows that evolve from notebooks into clean, maintainable modules

• Applied regression, classification, clustering, and feature engineering
• Cross-validation, hyperparameter tuning, and model selection
• Evaluation beyond score-chasing: interpretability, calibration, and failure-mode awareness

• Prompting, embeddings, semantic retrieval, and retrieval diagnostics
• RAG workflows with grounding, guardrails, and evaluation-oriented design
• Tool-aware patterns for practical, reliable assistant behavior

• ELT/ETL principles, warehouse-friendly modeling, and dbt-style transformation layers
• Metric definitions, consistency checks, and data quality gates
• BI-ready marts that support scalable dashboards and stakeholder trust

• KPI frameworks: definition, grain, drill-down, and segmentation
• Executive-facing dashboards, operational monitoring, and business storytelling
• Semantic clarity, naming discipline, filters, controls, and interpretable reporting

• Workintech — 480 Hours
  Module 1: Data Analyst
  Module 2: Data Scientist & AI Pro

• Data Analysis School — Council of Higher Education (YÖK) — 300 Hours
  Consortium: BOUN • METU • MU • ITU
  Module 1: Panel Data Analysis
  Module 2: Artificial Intelligence & Machine Learning

• BAU Bright & Wissen Academy — 440 Hours
  Full-Stack Software Development Specialist Program

Tooling listed reflects technologies used across academic training, portfolio work, and hands-on analytics, ML, BI, GenAI, and delivery workflows.

• M.Sc. Management Information Systems (MIS) — systems thinking, business analytics, and data-driven strategy
• MBA-level coursework within MIS — Business Analytics, CRM, Big Data Management, Sustainable Management, Value Chain, Project Management, Research Methods, and Low-Code Development
• B.Sc. Mathematics & Science Education — quantitative reasoning and structured problem solving
• B.A. Sociology — human behavior, society, and analytical interpretation
• Science High School foundation — STEM-focused academic background

• PMI-aligned Project Management background with emphasis on scope, planning, risk, and communication
• Agile and iterative delivery mindset with clear milestones and stakeholder alignment
• High-responsibility volunteering in Search & Rescue, strengthening discipline, teamwork, and decision-making under pressure

• Structured thinking with practical execution
• Transparent assumptions, clean workflows, and reliable evaluation
• Clear communication for both technical and non-technical audiences
• Documentation-first, collaboration-ready delivery