Data, AI and Software
Dual Degree in Computer Science and Psychology at Ben-Gurion University, 2026
Data, AI and software. I build the pipeline, the model and the app, then report honestly how well it works.
Idan Akiva
B.Sc. Computer Science & B.A. PsychologyBen-Gurion University2026Price: one email
The live dashboard of the World Cup 2026 Predictor, 2026 Predictions page, captured October 2026.
Three Years of Production Data Work
Pipelines Handling Hundreds of Millions of Data Points on Peak Days
Before university: BI Developer and Product Analyst in the C4I Corps, IDF, then Data Analyst and BI Developer at NESS Technologies. His data systems cut reporting work from days to minutes and earned the Chief of Staff’s Innovation Award.
Work
Three things I built, with the numbers and where each one comes from.
World Cup 2026 Predictor
Live demo — Predicts every match of the 48-team 2026 World Cup, then plays the whole tournament 10,000 times to estimate each team’s chances.
- Cleaned 1930–2022 match data with four SQL scripts into one SQLite model.
- Built 16 time-decay features and trained an XGBoost classifier on 964 matches.
- 57.1% ± 2.1% accuracy in 5-fold cross-validation, against a 55.7% always-home-win baseline. A small gain, reported as one.
Python · SQL · SQLite · XGBoost · Streamlit
What2Wear
Finished — An iPhone widget and app that turns the weather where you are into what to wear and what to pack.
- About 1,400 lines of JavaScript, running in Scriptable on the phone.
- Hour-by-hour feels-like temperature and rain from the Open-Meteo API.
- Trips weeks ahead fall back to the same dates over the last five years.
JavaScript · REST API · Scriptable
Mental Health & Tech Hackathon
No public code — Team of four, May 2025. We designed an AI-integrated hub that keeps an infant’s medical files in one place and turns doctors’ recommendations into activities at home. The organizers asked us to keep developing it.
Experience
Three years of production data work before university: pipelines handling hundreds of millions of data points on peak days, feeding reports used by thousands of people.
Data Analyst & BI Developer
- Built automated ETL pipelines that unify data from many sources into optimized structures, relied on by senior leadership for time-critical decisions.
- Turned operational requirements from cross-functional stakeholders into technical specifications and working reports.
BI Developer & Product Analyst
- Architected data systems that cut data retrieval and reporting work from days to minutes, earning the Chief of Staff’s Innovation Award.
- Owned the data product end to end: SQL and ETL backend, data modeling, Power BI dashboards and DAX measures, user testing, rollout and support.
- Built a telemetry data backend in Python, SQL and SSIS for a driver-assistance system, in a three-person team over a year and a half.
- Ran A/B comparisons between solution versions and chose what to roll out from the results.
- Learned SSIS and BI in weeks, then trained about seven developers and mentored juniors on database design and query optimization.
Figures as stated on my CV. The systems themselves are internal and not public.
Study
B.Sc. Computer Science and B.A. Psychology, Ben-Gurion University of the Negev, 2023–2026. GPA 84.8. The courses I did best in:
- Research Methodology
- 98
- Mathematical Applications in CS
- 96
- Regression & Multivariate Analysis
- 94
- Statistical Inference
- 92
- Applied Machine Learning
- 86
- Linear Algebra
- 86
Grades as listed on my CV.
Skills
Hebrew (native) · English (fluent)
Data & BI
- SQL
- ETL pipelines
- Data modeling
- SSIS
- Power BI and DAX
- SQLite
ML & statistics
- scikit-learn
- XGBoost
- pandas and NumPy
- Cross-validation
- Regression
- Monte Carlo simulation
- Jupyter
AI
- LLM API integration
- Claude Code
- Claude
- Gemini
Code & tools
- Python
- Java
- SQL
- JavaScript
- C and C++
- Git and GitHub Actions
- Streamlit
Contact
Email is the fastest way to reach me.