Stack
What I actually build with
Counted from the fifteen projects on this site rather than listed from memory. A tool is here because a project used it, and the number is how many did. Every one links straight to that work.
- Tools
- 50
- Used more than once
- 8
- Projects
- 15
Reached for repeatedly
- Python10×
- German Job Market Intelligence Corpus
- U.S. Flight Delay Analysis
- Quantifying Data Quality
- Does Environmental Performance Drive Economic Growth?
- Manufacturing Cost Variance Dashboard
- E-Commerce Return Rate Analysis
- Ganymede
- Witness
- Rainfall Estimation via Heterogeneous Data Fusion
- Fintech Real Estate Intelligence System
- Vercel6×
- pandas4×
- scikit-learn4×
- TypeScript4×
- Next.js3×
- LLM orchestration2×
- Supabase2×
Used once, on purpose
Picked for one job because it was the right tool for that job. Listing them separately is more honest than padding the count above.
- BeautifulSoup
- BigQuery
- Business intelligence
- Calibration
- Cohort analysis
- Computer vision
- Data modelling
- Decision tree
- Firebase
- GPT-4
- JavaScript
- K-Means
- KPI design
- LangChain
- LightGBM
- Llama 2/3
- Local-first state
- Logistic regression
- Meta-learner stacking
- Meteosat
- Microservices
- Next.js 15
- NumPy
- OLS
- Polars
- Postgres RLS
- Prisma
- Qlik Sense
- R
- Random Forest
- React
- REST APIs
- spaCy
- Spatiotemporal analysis
- SQL
- Statistical testing
- statsmodels
- Tableau
- TF-IDF
- Uplift modelling
- Welch's t-test
- Window functions