Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →World-class data science skill for statistical modeling, experimentation, causal inference, and advanced analytics. Expertise in Python (NumPy, Pandas, Scikit-learn), R, SQL, statistical methods, A/B testing, time series, and business intelligence. Includes experiment design, feature engineering, model evaluation, and stakeholder communication. Use when designing experiments, building predictive models, performing causal analysis, or driving data-driven decisions.
.claude/skills/davila7-senior-data-scientist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 927% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 214% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 154% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 63% | 0% |
World-class senior data scientist skill for production-grade AI/ML/Data systems.
bash# Core Tool 1 python scripts/experiment_designer.py --input data/ --output results/ # Core Tool 2 python scripts/feature_engineering_pipeline.py --target project/ --analyze # Core Tool 3 python scripts/model_evaluation_suite.py --config config.yaml --deploy
This skill covers world-class capabilities in:
Languages: Python, SQL, R, Scala, Go ML Frameworks: PyTorch, TensorFlow, Scikit-learn, XGBoost Data Tools: Spark, Airflow, dbt, Kafka, Databricks LLM Frameworks: LangChain, LlamaIndex, DSPy Deployment: Docker, Kubernetes, AWS/GCP/Azure Monitoring: MLflow, Weights & Biases, Prometheus Databases: PostgreSQL, BigQuery, Snowflake, Pinecone
Comprehensive guide available in references/statistical_methods_advanced.md covering:
Complete workflow documentation in references/experiment_design_frameworks.md including:
Technical reference guide in references/feature_engineering_patterns.md with:
Enterprise-scale data processing with distributed computing:
Production ML system with high availability:
High-throughput inference system:
Latency:
Throughput:
Availability:
bash# Development python -m pytest tests/ -v --cov python -m black src/ python -m pylint src/ # Training python scripts/train.py --config prod.yaml python scripts/evaluate.py --model best.pth # Deployment docker build -t service:v1 . kubectl apply -f k8s/ helm upgrade service ./charts/ # Monitoring kubectl logs -f deployment/service python scripts/health_check.py
references/statistical_methods_advanced.mdreferences/experiment_design_frameworks.mdreferences/feature_engineering_patterns.mdscripts/ directoryAs a world-class senior professional:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 9,297 | 11,000 | +18% | 1 | 1 | 0% | 1,692 | 3,623 | +114% | 0 | 0 | — |
case-02 | fail→pass | 2,415 | 8,125 | +236% | 1 | 1 | 0% | 221 | 2,270 | +927% | 0 | 0 | — |
case-03 | fail→pass | 8,552 | 9,772 | +14% | 1 | 1 | 0% | 918 | 2,884 | +214% | 0 | 0 | — |
case-04 | fail→pass | 5,367 | 2,818 | -47% | 1 | 1 | 0% | 933 | 1,666 | +79% | 0 | 0 | — |
case-05 | fail→pass | 3,996 | 2,697 | -33% | 1 | 1 | 0% | 670 | 1,704 | +154% | 0 | 0 | — |
case-06 | fail→pass | 6,286 | 2,516 | -60% | 1 | 1 | 0% | 1,043 | 1,705 | +63% | 0 | 0 | — |
case-07 | fail→pass | 7,085 | 2,079 | -71% | 1 | 1 | 0% | 1,248 | 1,568 | +26% | 0 | 0 | — |
case-08 | fail→pass | 6,613 | 5,605 | -15% | 1 | 1 | 0% | 1,297 | 2,221 | +71% | 0 | 0 | — |
case-09 | fail→pass | 4,884 | 2,580 | -47% | 1 | 1 | 0% | 971 | 1,620 | +67% | 0 | 0 | — |
case-22 | fail→pass | 3,406 | 1,635 | -52% | 1 | 1 | 0% | 481 | 1,412 | +194% | 0 | 0 | — |
case-10 | fail→pass | 7,200 | 3,099 | -57% | 1 | 1 | 0% | 1,255 | 1,780 | +42% | 0 | 0 | — |
case-11 | fail→pass | 7,352 | 4,026 | -45% | 1 | 1 | 0% | 1,395 | 1,881 | +35% | 0 | 0 | — |
case-12 | pass→pass | 4,062 | 1,255 | -69% | 1 | 1 | 0% | 700 | 1,335 | +91% | 0 | 0 | — |
case-13 | pass→pass | 6,677 | 2,101 | -69% | 1 | 1 | 0% | 1,307 | 1,536 | +18% | 0 | 0 | — |
case-14 | pass→pass | 2,830 | 1,572 | -44% | 1 | 1 | 0% | 407 | 1,433 | +252% | 0 | 0 | — |
case-15 | pass→pass | 7,990 | 3,933 | -51% | 1 | 1 | 0% | 1,429 | 1,776 | +24% | 0 | 0 | — |
case-16 | fail→pass | 7,056 | 3,961 | -44% | 1 | 1 | 0% | 1,314 | 1,832 | +39% | 0 | 0 | — |
case-17 | fail→pass | 4,203 | 2,413 | -43% | 1 | 1 | 0% | 686 | 1,545 | +125% | 0 | 0 | — |
case-18 | fail→pass | 10,205 | 5,112 | -50% | 1 | 1 | 0% | 1,803 | 2,187 | +21% | 0 | 0 | — |
case-19 | fail→pass | 4,232 | 1,807 | -57% | 1 | 1 | 0% | 651 | 1,484 | +128% | 0 | 0 | — |
case-20 | fail→pass | 7,457 | 3,161 | -58% | 1 | 1 | 0% | 1,341 | 1,821 | +36% | 0 | 0 | — |
case-21 | pass→pass | 5,466 | 4,326 | -21% | 1 | 1 | 0% | 1,029 | 2,026 | +97% | 0 | 0 | — |
case-23 | fail→pass | 12,990 | 15,053 | +16% | 1 | 1 | 0% | 2,844 | 4,347 | +53% | 0 | 0 | — |
case-24 | pass→pass | 5,105 | 4,640 | -9% | 1 | 1 | 0% | 814 | 2,013 | +147% | 0 | 0 | — |
case-25 | fail→pass | 9,657 | 8,656 | -10% | 1 | 1 | 0% | 2,068 | 2,947 | +43% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted. The headline lift of +72 percentage points is the difference between those two pass rates over the 25 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.