8+
years in production analytics and ML
Profile
Danny Pichardo
Senior data scientist focused on applied AI, machine learning, statistics, healthcare analytics, and baseball modeling. I work best at the intersection of data quality, model behavior, and high-stakes operational decisions.
3
major healthcare organizations across payer, provider, and registry work
2
public AQUA registry publications and abstracts
10 / 10
fantasy 5x5 categories led in my baseball projection stack
Experience
What I have actually built.
Senior Data Scientist / Lead GenAI Data Scientist
Blue Health Intelligence · Baltimore, MD
Jul 2020 - Present
Lead applied GenAI and data science work across healthcare use cases where model quality, governance, and practical adoption all matter.
- Led LLM evaluation and model selection for healthcare and data science workflows, including real-world evidence and care-gap use cases
- Built a text-to-SQL assistant with RAG over schema metadata, data dictionaries, and internal documentation
- Developed data-analysis agents with guardrails, logging, and evaluation loops for safe use on sensitive datasets
- Implemented an internal agent harness for coding assistants and trained researchers, analysts, and data scientists on effective usage
- Created a longitudinal clinical journey representation and knowledge base for claims histories
- Fine-tuned Llama 3 8B for clinical-journey tasks and scoped datasets, evaluations, and cost models for larger training runs
Senior Data Scientist
NewWave Telecom & Technologies · Elkridge, MD
May 2019 - Jul 2020
Led data science work for value-based care reporting and provider-facing analytics products tied to advanced payment models.
- Owned the data science workstream for a provider-facing reporting dashboard
- Built Databricks ETL and measure-calculation pipelines with validation workflows
- Delivered predictive modeling POCs using logistic regression, XGBoost, RNNs, and claim embeddings
- Built Looker data models and optimized SQL to meet sub-4-second dashboard performance SLAs
- Partnered directly with UX, clinical SMEs, and engineering to translate healthcare requirements into usable systems
Statistician / Data Analysis Manager
American Urological Association · Linthicum, MD
Jan 2017 - May 2019
Statistical and computational lead for the AQUA Registry and external consulting work tied to urologic quality reporting and research.
- Led analytic work for AQUA, a national quality registry and reporting platform
- Designed analyses to validate EHR extraction and NLP pipelines for completeness and accuracy
- Built predictive models for recurrence and complications across urologic procedures
- Maintained large healthcare data assets and guided scalable data-lake implementation work
- Coauthored published work and abstracts on registry adoption, quality metrics, and contemporary urologic practice
Independent projects
Side work with real technical substance.
baseball-data
A modern baseball analytics repo covering ingestion, feature-store design, projection systems, and market-oriented backtesting.
- Integrated Statcast, Retrosheet, Chadwick, and Lahman into analysis-ready layers
- Built walk-forward predictive 5x5 projection models with Bayesian blending and interval calibration
- Developed a repo-native prop betting framework for pricing, edge calculation, and bankroll-aware backtests
Browser-based bat-ball contact simulator
Interactive simulation app for exploring how pitch shape, timing, attack angle, and bat speed change likely batted-ball outcomes.
- 3D pitch-flight and bat-path modeling
- Rigid bat-ball collision model
- Monte Carlo perturbations for sensitivity and landing-spread analysis
Technical toolkit
Breadth with a point of view.
GenAI / LLM systems
Evaluation and benchmarking, RAG, analytics assistants, tool use, workflow design, fine-tuning, and human-in-the-loop deployment.
Statistics and modeling
Regression, predictive modeling, model comparison, calibration, walk-forward validation, causal thinking, measurement, and error analysis.
Data and platform stack
Python, SQL, Spark, Databricks, Snowflake, Azure, AWS, MLflow, Tableau, Looker, and production-minded data workflows.
Domain depth
Claims data, registry analytics, quality measures, real-world evidence, episode-based risk adjustment, healthcare operations, and baseball analytics.
Selected publications and talks
Public-facing work tied to my healthcare analytics background.
Characteristics of Participants in the American Urological Association Quality (AQUA) Registry and Early Impact of Participation on Quality of Care
Coauthored study on registry adoption and quality-measure performance in urology practice settings.
A Snapshot of Contemporary Urologic Practice From the American Urological Association's Quality Registry (AQUA)
Coauthored abstract summarizing national-scale practice patterns drawn from the AQUA registry.
Enabling Value-Based Health Care Using Modern Analytics Tools
Talk on modern healthcare analytics infrastructure, provider reporting, and the realities of building data products for value-based care.
Education
Formal training.
University of Maryland, Baltimore County
B.S. in Statistics · December 2014
Working style
What collaborators can expect.
- Start with the decision and the measurement frame, not the model pitch.
- Use advanced methods when they improve the work, not when they merely decorate it.
- Prefer transparent tradeoffs, strong baselines, and evaluation that survives deployment reality.
- Build systems people can trust enough to actually use.
Contact
AI / ML / statistics / healthcare analytics
For applied AI, machine learning, healthcare analytics, or statistical systems work, reach me at pichardoda@gmail.com.
Email
pichardoda@gmail.com
GitHub
github.com/DannyPi