Skip to content

Research & writing

Research interests

Beyond shipping systems, I'm interested in how AI changes the way teams work and decide. My focus sits at the intersection of agentic systems, multimodal retrieval, edge reliability, and enterprise adoption — grounded in production constraints rather than benchmarks alone.

Human–AI Collaboration & Workflow Redesign

Studying how AI-assisted planning reshapes product and engineering workflows — where humans stay in the loop, how trade-offs are surfaced, and how organizations adopt agentic decision support without losing accountability.

Enterprise AI for Sprint Planning

Continuous & Multimodal Video Retrieval

Exploring compressed spatio-temporal embeddings, multimodal fusion, and retrieval-backed LLM reasoning to make hours of video searchable, explainable, and useful for real-time decision support.

Video-RAG for Continuous Video Search

Edge AI for Safety-Critical Systems

Investigating how agentic and computer-vision pipelines can run dependably on edge hardware in safety-critical environments, balancing accuracy, latency, energy, and fail-safe behavior.

Production Edge Agentic AI for Mining Safety

Enterprise AI Adoption

Looking at how enterprises encode goals, constraints, and domain context so AI systems produce decisions that teams can trust, audit, and act on — and how to measure the resulting business impact.

Enterprise AI for Sprint Planning