Research
Published Work
Peer-reviewed and preprint research powering the Qualixar platform — AI agent reliability, behavioral contracts, supply chain security, and formal verification.
Varun Pratap Bhardwaj
→ 1,980 experiment sessions across 7 models
Read on arXiv →Varun Pratap Bhardwaj
→ 675 tests, 8 novel contributions
Read on arXiv →Varun Pratap Bhardwaj
→ 74.8% on LoCoMo (zero cloud) — highest local-first score reported
Read on arXiv →Varun Pratap Bhardwaj
→ Local-first architecture, Bayesian trust defense against memory poisoning
Read on arXiv →Varun Pratap Bhardwaj
→ Stochastic testing across non-deterministic agent workflows
Read on arXiv →Research Interests
Agent Behavioral Contracts
Formal specification and runtime enforcement of agent behavior
AI Agent Safety
Preventing drift, hallucination, and unauthorized actions
Supply Chain Security
Verifying and securing AI agent skill ecosystems
Formal Methods for AI
Mathematical verification of stochastic agent systems
Agent Testing
Stochastic testing frameworks for non-deterministic agents
Enterprise AI Deployment
Production-grade patterns for agent reliability at scale