Hi, I'm

Mylene Aradani

A final-year Applied Computer Science student at UCLL. I build across the stack — machine-learning pipelines, data analysis, full-stack web apps and cloud services on Azure — and I enjoy explaining the technology as much as building it.

Mylene smiling on a rooftop terrace overlooking Brussels

From the classroom to code.

I first trained as a secondary-school English teacher. During my teaching internship I learned to prepare, present and adapt an explanation to very different people. Then I switched to IT, and found that the same skill matters everywhere: an app, a dashboard or a model is only useful when people understand it.

My projects cover a broad mix: UrbanHeat.be, an individual machine-learning project from raw satellite data to a dashboard; Seriesly, a full-stack Spring Boot and Next.js app; PackNGo, a cloud-native app on Azure; and Linux server automation with Bash.

  • All-round IT profile
  • Honest about what I know
  • Clear communicator
  • Learns unfamiliar codebases

Selected projects

Individual projectMar to May 2026

UrbanHeat.be

Predicts and explains urban heat anomalies across five Belgian cities, from citizen-science weather stations and Sentinel-2 satellite imagery to an interactive dashboard.

  • Engineered geospatial features (NDVI, NDBI, SWIR, humidity, altitude, wind, solar radiation) with rasterio.
  • Compared Random Forest, XGBoost, a stacked ensemble with a Ridge meta-learner, and a PyTorch CNN on 16×16 satellite patches trained on a GPU in Google Colab.
  • Validated with 5-fold and Leave-One-Station-Out cross-validation (LOSOCV RMSE 0.575 ± 0.460), which revealed a spatial-leakage gap of about 18%.
  • Explained predictions with SHAP and plain-language summaries from Llama 3.3 70B via the Groq API.
PythonPyTorchscikit-learnXGBoostrasterioSHAPGroq API
155
weather stations
5,575
station-day observations
0.599
R² of the stacked RF + XGBoost model
~18%
spatial-leakage gap found

Pair project, I did the implementationSep to Dec 2025Full-stack

Seriesly

A web app to discover and organise Asian dramas, movies and variety shows, with personal watchlists and an admin dashboard.

  • Search and filters, show and actor detail pages, and watchlists with Watching, Completed and Plan to Watch.
  • Admin dashboard for managing shows and actor information.
  • REST API in Java 21 and Spring Boot, secured with Spring Security, JWT and an OAuth2 Resource Server.
  • PostgreSQL with Spring Data JPA and Hibernate; tested with JUnit and Jest; documented with Swagger / OpenAPI.
Next.jsReactTypeScriptTailwind CSSJava 21Spring BootSpring SecurityPostgreSQLJUnitJest

Group projectFeb to May 2026

PackNGo

A cloud-native travel-planning app on Microsoft Azure.

  • Developed backend features in Java / Spring Boot and frontend features in React / Next.js.
  • Integrated Redis caching; worked on event-driven features with Azure Service Bus and Azure Functions.
  • Worked on the architecture and the Azure deployment; a teammate set up the GitHub Actions CI/CD pipeline.

Cosmos DB was set up by a teammate; I worked with it.

JavaSpring BootReactNext.jsRedisAzure FunctionsService Bus
The Dynamic Workflow Engine team posing in front of a Christmas tree
Our team. I'm second from the left.

Team of 7, junior developerSep 2025 to Jan 2026

Dynamic Workflow Engine

A modular app for dynamically defined business processes. A senior teammate built the core engine; I worked around it.

  • Worked across MongoDB, backend and frontend in a Java / Spring Boot codebase.
  • Modified and tested existing workflow logic within an event-driven architecture.
  • The architecture was hard to read at first; I kept at it until I could work across components.
JavaSpring BootMongoDBRESTTesting

More projects

OH Leuven attendance predictionTeam of 6, real client, Mar 2026

A machine-learning model predicting match-day attendance (tickets scanned) for 71 OH Leuven home matches.

  • Contributed to feature engineering from ticket sales (B2C/B2B), promotions, free-ticket share, Google Trends interest and press-article counts.
  • Contributed to feature selection (RFECV) and model comparison with Leave-One-Out cross-validation.
  • The team's best model, Gradient Boosting, reached R² 0.677 and an MAE of about 870 spectators.
Pythonpandasscikit-learnScrum
Linux server administration and automationGroup project, Oct to Dec 2023
  • Automated nightly server backups with a Bash script (input validation, logging, 7-day retention) scheduled via cron.
  • Wrote a Bash script that creates Linux users from a CSV, and deployed the team website on Apache with SSH-key access for five members.
  • Analysed Telnet vs SSH traffic in Wireshark and simulated the network in Packet Tracer.
LinuxBashcronApacheSSHWireshark
YINSH and UNO gamesTeam projects

Digital versions of two games in Java and JavaFX. I designed relational tables and wrote SQL queries and joins.

Home network setupPersonal, Apr 2024

Configured a Ubiquiti access point and three cameras, administered the connected devices and troubleshot connectivity and access issues.

Challenges and hackathons

Competitions and hackathons where I tested what I know against real problems and a deadline.

Individual competitionSep to Nov 2026

CIO Challenge by BNP Paribas Fortis

An IT challenge run on EDITx. I reached the finals and was invited to the final event in Brussels on 17 November 2026.

50

finalist out of 688 participants

Oct 2026, Leuven

Team of 4Data for Good Challenge 2026

MSF Field Library

Organised by Emergent Leuven with Médecins Sans Frontières Belgium as challenge partner.

  • MSF's question: how can field staff quickly find medical protocols during a humanitarian crisis, when internet is slow or gone?
  • Our answer: a search tool that runs fully offline on a CPU, over 564 MSF documents in English, French, Spanish and Arabic.
  • It answers in about 150 ms with hybrid search and quotes MSF's own documents word for word, so nothing is made up. A graph shows related documents and translations.
Hybrid search4 languagesOffline, CPU only

Nov 2026, online

Team of 3NASA Space Apps Challenge 2026

Flame in Freefall

NASA is planning a Moon base, and fire behaves very differently in low gravity.

  • We built an interactive, AI-powered web dashboard that summarises, ranks and interprets NASA's PSI combustion-science investigations.
  • It shows how much of NASA's space fire data was gathered in Moon-base air: of 659 tests that state their air, only 20 match it.
  • A habitat view lets you pick the air and materials and see what NASA has tested for each.
Data visualisationNLPLLM summaries

Skills

Every skill shows where it comes from.

  • Built it myself
  • Built in a team
  • Learning in current courses

AI and machine learning

  • Python
  • PyTorch
  • scikit-learn
  • XGBoost
  • CNNs, computer vision
  • SHAP, explainable AI
  • LLM API integration
  • Cross-validation
  • Transfer learning

Data analytics

  • Feature engineering
  • Geospatial data, rasterio
  • Model evaluation
  • pandas
  • Power BI
  • D3.js
  • R

Software development

  • Java
  • Spring Boot
  • REST APIs
  • Spring Security, JWT
  • React, Next.js
  • TypeScript
  • PostgreSQL, JPA
  • JUnit, Jest
  • MongoDB
  • Git, GitHub, Scrum

DevOps and cloud

  • Azure Functions
  • Azure Service Bus
  • Redis
  • CI/CD (exposure)
  • Linux, Bash
  • Docker (exposure)
  • Kubernetes, k3s
  • K9s, Lens
  • GitOps
  • Infrastructure as Code

Experience

Work where I learned to coach, present and look after details.

UCLL students on campus stairs at an ambassador event Collage of UCLL ambassador activities in Leuven
  1. Nov 2024 to now

    UCLL programme ambassador

    I represent my programme and inform prospective students, presenting and talking with people from very different backgrounds.

  2. Jul 2025 to Sep 2026

    Team member, then team coach at KFC

    Promoted to team coach in January 2026, training and coaching new team members in a fast-paced, customer-facing job. Selected for shift-leader training (Jul to Sep 2026), then ended the contract to focus on my studies.

  3. May to Dec 2021

    Office assistant, Mayor's Office, Malita

    Maintained records and Excel-based data files; drafted and proofread official correspondence, reports and memos.

  4. 2019 to 2023

    Teaching internship

    Prepared lessons, presented, adapted explanations to different learners and gave feedback.

Education

  1. 2024 to 2027

    UC Leuven-Limburg (UCLL)

    Professional Bachelor of Applied Computer Science, final year

  2. 2023 to 2024

    Karel de Grote University of Applied Sciences and Arts

    Applied Computer Science with a specialisation in AI, one year

  3. 2019 to 2023

    Southern Philippines Agri-Business and Marine and Aquatic School of Technology

    Bachelor of Secondary Education, major in English, completed

Certifications and badges

All my badges are on my Credly profile.

This semester I'm taking Cloud & Operations, AI Applications, and Data Visualisation & Analytics.

Let's work together.

I'm looking for a curriculum internship in Belgium from late January to May 2027, for at least 52 days, in AI, data analytics, software development, DevOps or cloud. I'd love to hear about your team.

Happy to work in Leuven, Brussels, Ghent, Antwerp, Mechelen or Mol.