Skip to content
View Jevintlie's full-sized avatar
1 follower · 0 following

Block or report Jevintlie

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Jevintlie/README.md

Jevint Felixciano

Computer Science graduate building practical software across applied AI, computer vision, machine learning, and interactive systems.

I enjoy turning ambiguous requirements into focused products: defining the system, implementing the core workflow, testing the risky parts, and presenting the result clearly. The projects below are refined editions of university and self-directed work, with private material, generated output, and unlicensed assets removed.

Applied AI and data

  • University Programme RAG MCP Server — A source-linked retrieval service that turns programme pages into validated records, ChromaDB indexes, and MCP search tools. Includes deterministic ingestion, JSON Schema validation, offline tests, and configurable data paths.
  • Student Dropout and Academic Success Prediction — A leakage-aware multiclass benchmark using class-weighted and resampled pipelines. The repository documents dataset provenance, reproducible evaluation, and the limits of using educational data for real decisions.
  • Privacy-Aware Video Processing Pipeline — A reusable OpenCV command-line pipeline for enhancement, face anonymization, compositing, transitions, and branded output. Refactored from group coursework into tested image operations without publishing private footage.

Software and product work

  • Snakie — A ScalaFX desktop game with separated game logic, score persistence, difficulty modes, automated tests, and a documented architecture.
  • Zephyrus G14 Product Study — A responsive, accessible product microsite rebuilt from foundational web coursework using semantic HTML, modern CSS, vanilla JavaScript, and original CSS artwork. Source
  • Smart Vehicle Performance System — A product and system-design case study connecting requirements, UX decisions, vehicle telemetry, cloud services, analytics, and UML models. The repository clearly separates my leadership and individual artifacts from shared group work.

Working principles

  • Build the smallest system that demonstrates the idea honestly.
  • Keep domain logic testable and external dependencies replaceable.
  • Treat accessibility, privacy, attribution, and reproducibility as engineering requirements.
  • Document limitations and trade-offs instead of hiding them behind presentation polish.

Technical focus

Python · Scala · JavaFX / ScalaFX · JavaScript · HTML / CSS · OpenCV · scikit-learn · XGBoost · ChromaDB · MCP / JSON-RPC · UML · Git

Pinned Loading

  1. rag-mcp-server rag-mcp-server Public

    Source-linked RAG MCP server for structured university programme information.

    Python

  2. ROG ROG Public

    Responsive 2020 Zephyrus G14 product study rebuilt from foundational web coursework.

    CSS

  3. student-dropout-ml student-dropout-ml Public

    Leakage-aware multiclass benchmark for student dropout and academic-success prediction.

    Python

  4. opencv-video-pipeline opencv-video-pipeline Public

    Privacy-aware OpenCV pipeline for enhancement, anonymization, compositing, and branded video output.

    Python

  5. Snakie Snakie Public

    ScalaFX desktop snake game with separated game logic, persistence, tests, and portfolio documentation.

    Scala

  6. smart-vehicle-performance-system smart-vehicle-performance-system Public

    Product design and UML case study for a connected vehicle performance and maintenance platform.