EJF.
Software Engineer — Brussels, BE

Elie-Joe
Farah

I build backends, databases, and data systems, and I'm moving into renewable energy research. Currently a Master's student in computer science at the Université Libre de Bruxelles.

Elie-Joe Farah
50.8503° N, 4.3517° E
Role
Software Engineer
Focus
Systems / Data
Research
Renewable Energy
Based
Brussels, BE
Status
Open to work

About

01About

I'm a software engineer working on backends, data modeling, and database systems, with a growing focus on renewable energy research. My BE is in Computer Engineering from the Lebanese American University (Dean's Honor List), and I'm now finishing a Master's in Computer Science & Engineering at the Université Libre de Bruxelles.

So far I've worked across QA and testing, backend APIs, database internals, and analytical platforms — through three internships in Beirut and a set of personal projects. Outside work I read maps and walk them: cartography and hiking are how I spend most of my time away from a screen.

Software

JavaPythonC++ ReactNode.jsPostgreSQL Machine Learning

Hardware & Systems

System DesignVHDL VerilogFPGAAssembly

Experience

02Experience
Jun 2025 — Jul 2025

QA Intern

FOO · Beirut
  • Ran 2,500+ mobile and web test cases and caught regressions before they reached production.
  • Filed 25+ critical bugs in Jira, improving coverage and workflow clarity.
Jun 2024 — Aug 2024

Software Dev & QA Intern

ITG · ITB · Beirut
  • Built a PHP/Laravel backend demo with a REST API and authentication.
  • Did data-driven testing and found 20+ major issues.
Jun 2023 — Aug 2023

Hospital Info System Intern

Hôtel-Dieu de France · Beirut
  • Wrote SAP/Oracle SQL reports for clinical analytics and optimized the queries behind them.

Projects

03Selected Work
P—01
Full-stack · Energy

BatteryLCATool

A multi-tenant platform for battery life-cycle inventory modeling. Manufacturers enter process inventories by hand or via Excel import; admins map flows to the ecoinvent database and export VUB-format workbooks. It includes a Monte Carlo engine with parameterized uncertainty distributions and Spearman correlation sensitivity analysis.

FastAPIPostgreSQLSupabaseReactTypeScriptNumPy/SciPy
View project →
P—02
Database internals · C

PostgreSQL DNA Extension

A PostgreSQL extension in C that adds native DNA sequence types with bit-packed storage (2 bits per nucleotide) and k-mer substring indexing through an SP-GiST trie. You get typed DNA columns, alignment queries, and index-accelerated k-mer lookups directly in SQL.

CPostgreSQLDatabase Internals
View source →
P—03
Compiler · Java

yaLcc Compiler

A compiler for a C-like language, built from scratch in Java. Lexical analysis with JFlex feeds a recursive descent parser, a symbol table with type checking, and IR code generation. Handles functions, control flow, arrays, and structs.

JavaCompiler DesignJFlex
View source →
P—04
EdTech · Applied AI

AI Course Platform

An educational platform using generative AI to create personalized course content and quizzes. Deployed and tested in a real classroom, where environmental and interaction data were collected through sensors on a Raspberry Pi and sent over a REST API, enabling context-aware learning experiences.

ReactNode.jsOpenAI APIRaspberry PiREST API

Research

04Research
2026 · Submitted
REWIND: European Wind Fleet LCA
E-J. Farah, co-authors. International Journal of Life Cycle Assessment.
A co-authored study introducing REWIND, a life-cycle assessment model that estimates the environmental impact of every wind turbine installed in Europe (77,552 of them) from each turbine's own size and location rather than country-level averages. Rebuilt the model's impact engine on the lca_algebraic library, which the earlier model didn't use: instead of re-solving a large matrix for every single turbine, impacts are written as formulas that each turbine's parameters plug straight into, cutting evaluation to a fraction of a second per turbine and turning a previously impractical computation into a workable fleet-wide run. Also handled data curation and the geospatial pipeline: assembling the turbine register, deriving grid-connection cable lengths from OpenStreetMap substation data, and estimating per-turbine electricity output from wind-atlas measurements.
Pythonlca_algebraicOpenStreetMapNumPy / PandasLife Cycle Assessment
Submitted
05 Lebanon 33.85° N, 35.86° E
Hiking in Lebanon
On foot · Mount Lebanon
Hiking in Lebanon
On foot · Mount Lebanon
Hiking in Lebanon
On foot · Mount Lebanon
Hiking in Lebanon
On foot · Mount Lebanon

Contact

06Contact
Get in touch

I'm looking for new opportunities. Email is the best way to reach me.