safety-analysis

The Manifold Alignment Protocol (MAP) is a geometric analysis framework for studying how complex systems—such as large language models, diffusion models, or physical measurement pipelines—converge, stabilize, and align under iterative processes.

A multi-agent AI system using Google AI Agent SDK and Gemini to analyze real-time route safety (crime, weather, lighting) and optimize navigation.

Real-time ADAS safety analysis using RF-DETR vs YOLOv8 across 7 adverse driving scenarios. Physics-based stopping distance model with ISO 22179 methodology.

Tool for visualizing requirements and generating model skeleton

ISO 26262 compliant ECC safety analyzer for systems like LPDDR4/5

Hierarchical imprecise reliability assessment and failure propagation for LLM-based agentic systems: LangGraph to fault trees to Bayesian networks

AI-powered image safety analyzer using BLIP + LLaMA 3.3 via Groq API

A business intelligence solution that transforms occupational health and safety data into actionable insights using Power BI. Features interactive dashboards for incident analysis, risk assessment, PPE compliance, LTIFR/TRIFR monitoring, and executive safety reporting to drive proactive decision-making.

Theoretical review of physics-guided multimodal strategies for WHO Grade IV tumours, covering imaging, radiotherapy, ablation, hyperthermia, TTFields, safety limits, ethics, and future oncology research.

Architecture-based Risk Analysis