Pinned Loading

1. auditable-multi-agent-recommender auditable-multi-agent-recommender Public

   Auditable multi-agent recommender for cloud optimization. Orchestration beats single-shot frontier models on correctness (18/18 vs 3–4/18), with a fully replayable evidence trail. Human-in-the-loop.

   Python

2. llm-wiki-knowledge-engine llm-wiki-knowledge-engine Public

   Three-layer hybrid RAG over an LLM-curated wiki + knowledge graph, with corpus-as-ground-truth evaluation. Applies Karpathy's LLM Wiki concept to ITSM tickets. (In progress)

   Jupyter Notebook

3. movie-review-sentiment-service movie-review-sentiment-service Public

   MLOps stack around a fine-tuned DistilBERT served in-JVM via Deep Java Library — Kotlin/Ktor microservices, RabbitMQ async, Prometheus + Grafana observability. CSCA 5028 capstone (CU Boulder MS-CS).

   Kotlin