Work

Things I have built and studied.

Featured case studies first, then public code and course experiments. Every item is labeled by type, so it is clear what was personal, team, or coursework.

01Personal projectWorking prototype

CareerGENAILLM Career Mentor

A full-stack AI career mentor that reads résumé inputs, identifies skill gaps, and turns them into a learning roadmap.

Problem Career advice is usually generic. A student can name a target role but rarely knows which gaps matter most or what to build next.

  • Next.js
  • FastAPI
  • Gemini 2.5 Flash

02Team capstoneCompleted capstone · platform concept

BECAYISCivil Servant Relocation Matching

A software engineering capstone exploring how optimization could support relocation matching for Turkish civil servants.

Problem In Türkiye, civil servants can relocate by swapping posts with a colleague who wants the reverse move, a process called becayiş. Finding a compatible partner by hand is slow, and the search gets harder once preferences and constraints stack up.

  • React
  • Spring Boot
  • PostgreSQL

03Team course projectIn progress · Fall 2026

Robotic ReceptionistCampus Human–Robot Interaction

A campus receptionist prototype that explores how people interact with a robot when asking for directions or student services.

Problem Visitors to a university lobby ask the same questions again and again: where an office is, which floor a department is on, who can help with a student service. Staff are not always at the desk.

  • Computer vision
  • HRI
  • Python

Public code

Pinned repositories on my GitHub.

All repositories on GitHub (opens in a new tab)

Selected experiments

Course labs where I practiced a method end to end. They are smaller in scope than the case studies and use course or public datasets.

  • Course project · NLP

    Quora question topic modeling

    TF–IDF · NMF

    Extracted 20 interpretable topics from a large Quora question corpus with TF–IDF features and non-negative matrix factorization.

  • Course project · Deep learning

    FRED time-series forecasting

    LSTM · sliding windows

    Trained an LSTM on economic series from FRED using 12-step input windows to forecast the next value.

  • Course project · Evaluation

    Supervised model comparison

    Logistic regression · KNN · SVM · tree ensembles

    Compared a logistic regression baseline with KNN, SVM, decision trees, random forests, and boosting, using macro F1 so minority classes counted.

  • Course project · Databases

    MySQL partitioning and fragmentation

    RANGE · LIST · HASH · EXPLAIN

    Partitioned a synthetic 100,000-row table three ways, checked partition pruning with EXPLAIN, and designed vertical fragmentation across two AWS Regions.

  • Course project · Databases

    Graph queries in ArangoDB

    AQL traversal · shortest path

    Wrote graph traversals for recommendations and shortest paths, and compared the model with a document store in MongoDB.

  • Course project · Cloud

    AWS storage lab

    EBS · snapshots · S3 versioning

    Resized an EBS volume and its filesystem, used incremental snapshots, restored S3 object versions, and set lifecycle rules with cost estimates.