I design and ship production LLM systems — agentic platforms, hybrid RAG, multi-agent orchestration, and the evals that keep them honest.
- Washington DC-Baltimore Area
-
01:53
(UTC -04:00) - in/uehlingeric
Pinned Loading
-
agentic-rag
agentic-rag PublicProvider-agnostic agentic RAG reference system: multi-agent orchestration, hybrid retrieval, guardrails, judged evals
Python
-
federal-llm-blueprint
federal-llm-blueprint PublicTerraform reference architecture for federal LLM workloads: no-egress VPC, KMS everywhere, NIST 800-53 mapping
HCL
-
banana-mcp
banana-mcp PublicMCP server exposing Gemini image and text generation as Claude Code tools; single and batch, persistent defaults
Python
-
dog-breed-classification
dog-breed-classification PublicCNN-based dog breed classifier achieving 95% accuracy across 70 breeds using ensemble learning
Jupyter Notebook
-
ground-combat-systems-contract-analysis
ground-combat-systems-contract-analysis PublicDoD ground combat systems contract analysis (FY2016–2020); vendor concentration, lifecycle, risk assessment
Jupyter Notebook
-
interpretable-diabetes-prediction
interpretable-diabetes-prediction PublicXGBoost diabetes prediction; 96% ROC AUC on Kaggle EHR data; clinical vs. survey data comparison
Jupyter Notebook
If the problem persists, check the GitHub status page or contact support.