Research | Sasan Jafarnejad | AI Researcher

Agentic & Multimodal AI

Mon, 01 Jan 0001 00:00:00 +0000

I study architectures that combine multimodal perception with planning, retrieval, and tool use. The goal is to build agents that can ground their reasoning in real environments and make their decisions more transparent and dependable.

This work includes multimodal large language models, retrieval-augmented generation, multi-agent systems, and evaluation methods for real-world tasks.

Intelligent & Human-Centered Mobility

Mon, 01 Jan 0001 00:00:00 +0000

Mobility is a demanding environment for intelligent systems: decisions are contextual, safety matters, and human needs are often ambiguous. I investigate how agentic vehicles, conversational interfaces, and digital twins can improve transportation while keeping people at the center.

Current themes include human–vehicle collaboration, geospatial reasoning, passenger safety, and communication-aware fleet intelligence.

Applied AI Systems

Mon, 01 Jan 0001 00:00:00 +0000

I build and evaluate end-to-end AI systems rather than treating models in isolation. This includes information retrieval, private on-device inference, data-quality agents, and decision-support tools in domains such as mobility and financial compliance.

The recurring question is practical: what must change for a promising model to become a useful, reliable system?