Independent Researcher | Exploring human-centered AI, cognitive systems, computer vision, and computational approaches to understanding complex patterns
🏢 Senior AI/ML Engineer at Nutanix | 15+ years in software engineering, building LLM, RAG, agentic, and graph-based systems for system intelligence

• AI Systems: LLM applications, RAG and GraphRAG, agentic workflows, structured extraction, knowledge graphs
• Machine Learning & Computer Vision: PyTorch, Hugging Face Transformers, DINOv2, OpenCV, visual embeddings, representation learning, image retrieval
• Human-Centered & Cognitive AI: Cognitive modeling, temporal and graph-based reasoning, affective computing, explainability
• Engineering & Infrastructure: Python, FastAPI, Pydantic, Apache Spark, Neo4j, Docker, Ollama, AWS

BeliefTrace
An experimental cognitive AI research project for extracting structured belief representations from narrative text and tracing their temporal evolution. The framework consists of two core modules:

• Belief Model: A QLoRA/PEFT-based model for multi-label core-belief extraction.
• Longitudinal Belief Graph: A temporal state machine for modeling how inferred beliefs persist, conflict with new evidence, and change state across sequential narratives.

Research Focus: Synthetic data augmentation, ablation studies, multi-label evaluation, hallucination tracking, confusion matrix analysis, and paired bootstrap significance testing.

Affective-Bridge
A middleware system for affective signal interpretation and affect-aware AI interaction, with a focus on privacy and explainability.

My research explores how machine learning can uncover and reason over complex patterns across visual, cognitive, affective, and temporal data.

I am currently exploring computer vision and computational iconography, including visual representation learning, motif retrieval, and fine-grained geometric discrimination, with longer-term interest in connecting visual evidence with historical context, knowledge graphs, and grounded AI reasoning.

In parallel, I continue exploring human-centered and cognitive AI, including affective modeling, longitudinal representations of beliefs and internal state, and graph-based approaches to understanding how patterns evolve across time and context.

Across these directions, I am particularly interested in systems that combine representation learning, structured knowledge, explainability, and human interpretation rather than treating perception or reasoning as isolated problems.