Over Thomas
Most companies don't have a data problem, they have a decision problem. Reports get built, dashboards opened, and actual choices still get made on instinct. I turn data into decisions your teams can act on.
Where your data likely loses value
- Reporting describing the past, not the next move: dashboards nobody acts on, KPIs disconnected from decisions. The reporting exists, the insight doesn't.
- Decisions made blind on things that are predictable: the signal is usually present in your data; nobody has modelled it, or the model was never trusted enough.
- Much effort and time spent on the plumbing (extracting, cleaning and reconciling data) instead of analysis).
My approach
- Business first, technology second: I start with your context, constraints and objectives. The analysis is scoped to a decision you have to make.
- Diagnostics: where numbers come from, their usefulness, which questions are answerable now and which need better data.
- Modelling: from descriptive analysis to predictive models - validated, interpretable, and calibrated to needs.
- Built to run: automated pipelines, reproducible outputs, documentation; delivered so your team can own it.
- Autonomous, from business question to business decisions.
Recent projects
- Quality score prediction (product data): random forest regression, driver identification and SHAP interpretation — 1.3% MSE on test set, enabling automated rating and classification.
- ETL pipeline & financial BI dashboard (SQL, Power BI): end-to-end: ingestion, modelling, dashboard generation for performance and risk monitoring.
- Hedge fund-grade systematic trading infrastructure (Python): data acquisition, DuckDB/parquet storage, custom backtesting engine and walk-forward optimization, deployed for live execution.
- Equity screening engine (production, daily runs): supervised classification across global equity markets, data acquisition, parallel computing, automated deployment.
Github: ThomasLaloux
Engels
Tweetalig / moedertaal
Frans
Tweetalig / moedertaal
Nederlands
Basiskennis
Kan op locatie werken
Brussels (tot 10km), Namur (tot 10km)
Werkervaring
- Self-employedData Analyst & Workflow/AI Automation Developerseptember 2025 - Vandaag (11 maanden)- Data analysis and predictive modeling (python): exploratory data analysis, calibration and deployment of a XGBoost predictive model for product quality assessment- Data analysis (SQL, Power BI): ETL pipeline and financial dashboard- AI/workflow automation (python, n8n): business-specific document generation under constraints - multi-agentic system integrating RAG, APIs, specialized prompts, GenAI (local SLM/cloud LLM), n8n orchestration, allowing to save 20-40h a month- Quantitative development (python): architectural design and implementation of a production-grade/scalable systematic trading infrastructure - custom backtesting engine, live trading, monitoring of performance & risk metrics
- Self-EmployedProprietary trading & quantitative developmentjanuari 2018 - november 2025 (7 jaren en 10 maanden)- Eight years running my own book and building the systems behind it, by choice, not as a client-service practice.Developments now largely complete, opening to client work in 2026.- Data analysis & equity market screening (matlab): performed data analysis and supervised classification (comprehensive screening engine) to identify position trading opportunities across global equity markets on a daily basis (data acquisition with APIs, trend following strategies, custom visualization, parallel computing, deployment in production.- Algorithmic trading (python, mql5): developed and backtested mid-frequency algorithmic trading systems (long/short trend following, mean reversion), coded advanced day trading indicators to support discretionary
- ENGIEBusiness & Market Analystjuli 2012 - december 2017 (5 jaren en 5 maanden)Brussels, Belgium- Data analysis (SQL, python): applied SQL queries and ML classification techniques to cluster of clients behavior from smart thermostat data and produce energy efficiency insights.- Power market modelling: modelled the LT Italian supply, demand and power prices as a key driver for strategic positioning and investment files.- Consulting in economic modelling (Excel): designed total cost of ownership models comparing green mobility solutions across energy types. Cooperated with M&A teams on assets valuations. Managed budgets, relationships with internal clients, and VIEs.- Market analysis: studied green mobility market potential, competition & regulation.
Aanbevelingen
Wees de eerste die Thomas aanbeveelt
Help deze freelancer om te schitteren door te vertellen hoe het is om met hem of haar te werken.
Deze freelancerprofielen matchen ook met zoekopdracht.
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Opleidingen
- Diploma of Advanced Studies in StatisticsUCLouvain2006Diploma of Advanced Studies in Statistics
- MSc. in StatisticsUCLouvain2005MSc. in Statistics