ResearchResearch

EdgeJury

Multi-LLM council with cross-review for truthful QA on serverless edge inference.

Read Paper

🎯The Problem

Single LLMs can hallucinate or miss edge cases. Large frontier models are expensive for edge deployment.

💡The Solution

Run questions through a council of role-specialized small models, force structured critique, rank candidates, and synthesize a final answer with consistency labeling.

✨Key Highlights

  • Cross-reviewed small-model ensembles
  • Structured critique + synthesis
  • arXiv published (2601.00850)

Other Projects

HyperSentry

An agentic security copilot that uses policy checks + retrieval to generate remediation PRs and accelerate incident analysis.

SchemaPulse

OpenAI-compatible inference layer for schema-valid structured outputs with streaming tool-call parsing.

TariffWhisperer

End-to-end RAG system over CBP rulings for explainable HTS code recommendations.