Henry Bernreuter

AI Engineer & Full Stack Data Engineer  ·  Atlanta, GA

LinkedIn  ·  GitHub  ·  Contact


Skills

  • Languages: Python, R, SQL, TypeScript
  • AI / ML: Anthropic Claude API, LangChain, RAG, Hugging Face Transformers, scikit-learn, TensorFlow, PyTorch, NLP, vector embeddings, agentic workflows
  • Data Engineering: Dagster, Snowflake, AWS S3, ETL/ELT pipelines, dbt, Bloomberg Data License, Aladdin XML, SDTM/ADaM, SAS
  • Backend: FastAPI, Node.js, Express, REST APIs, Redis, Celery, JWT authentication, PostgreSQL
  • Frontend: React.js, TypeScript, R Shiny, Plotly, Tableau
  • Cloud / DevOps: AWS EC2, Docker, Nginx, GitHub Actions, GitLab CI/CD, Linux, AWS IAM, Posit Connect

Experience

Triforai  ·  Full Stack AI Engineer

Atlanta, GA  ·  January 2025 – Present

  • Architected and deployed full-stack AI applications using Python, FastAPI, React.js, TypeScript, AWS EC2, Docker, Nginx, GitHub Actions, and REST APIs, integrating Generative AI and LLM workflows (Anthropic Claude API) with multi-turn conversation memory, secure user session isolation, and automated DOCX/PDF publishing pipelines.
  • Designed and implemented enterprise Data Engineering and Workflow Orchestration platforms using Dagster, Snowflake, AWS S3, ETL/ELT pipelines, YAML-driven asset factories, and distributed batch processing to automate Bloomberg, Aladdin, ELIC, GPIM, ADC, and financial transactions data pipelines across cloud-native architectures.
  • Engineered cloud-based AI and analytics infrastructure combining FastAPI microservices, React frontends, JWT authentication, CI/CD, GitLab, Linux, AWS IAM, Docker, Redis, Celery, and Snowflake to support audit trails, SCD Type 2 history tracking, real-time operational workflows, and scalable multi-tenant enterprise data platforms.
  • Designed and developed AI agent workflows using Python, FastAPI, LangChain, REST APIs, vector embeddings, and LLM orchestration patterns to automate multi-step reasoning, contextual retrieval, document generation, and user-driven decision support across clinical and enterprise data platforms.
  • Applied AI Engineering, NLP, Generative AI, scikit-learn, TensorFlow, and Hugging Face Transformers to support healthcare analytics, patient stratification, predictive modeling, and automated clinical insights generation across large-scale healthcare and RWE datasets.

BioMarin Pharmaceuticals  ·  Clinical Data Engineer

San Rafael, CA  ·  April 2021 – November 2024

  • Designed, developed, tested, and maintained SAS and R programs to support statistical analysis and reporting of clinical trial data in compliance with ICH guidelines and regulatory requirements.
  • Delivered high-performance data processing pipelines for SDTM and ADaM datasets, including generation of TFLs used in clinical study reports and regulatory submissions.
  • Deployed and managed Posit Connect for publishing Shiny apps, APIs, and reports, integrating authentication, TLS, external databases, and automated deployment pipelines in regulated environments.
  • Provisioned Linux-based cloud instances (AWS EC2) for Posit Workbench.
  • Developed oncology-focused AI and Clinical Data Management workflows using Python, R, SQL, TensorFlow, PyTorch, scikit-learn, and R Shiny to transform clinical trial, biomarker, and Real-World Evidence (RWE) datasets into analysis-ready data products supporting exploratory analytics and immunotherapy research.
  • Built interactive analytics applications and REST API integrations supporting clinical reporting, statistical programming, NLP-based text analysis, and AI-assisted healthcare workflows for cross-functional scientific and operational teams.

Purchasing Power  ·  E-Commerce Data Modeling & Pricing

Atlanta, GA  ·  January 2020 – January 2021

  • Built cloud-based customer analytics and pricing intelligence platforms using Python, SQL, R Shiny, Tableau, Snowflake, AWS, pandas, and machine learning workflows to analyze large-scale e-commerce, transactional, and behavioral datasets.
  • Developed predictive modeling and NLP workflows using TensorFlow, scikit-learn, Bayesian modeling, clustering algorithms, and recommendation systems to support customer segmentation, forecasting, and operational analytics.
  • Engineered scalable reporting and Data Engineering workflows integrating REST APIs, batch processing, Git, Linux, Docker, and CI/CD automation to support enterprise business intelligence and operational decision-making.

Education

Georgia State University  ·  Atlanta, GA

Master of Science, Information Systems  ·  May 2021
Concentration: Big Data Management and Analytics

Georgia State University  ·  Atlanta, GA

Bachelor of Science, Computer Science  ·  December 2019
Concentration: Game Design and Development


Projects

See the portfolio section of this site for selected project write-ups.


LinkedIn  ·  GitHub  ·  Contact