Quasi-Unitary State Space Model breaking the KV-cache scaling wall. Employs Lie-algebraic unitary state dynamics for true constant O(1) inference memory regardless of sequence length, enabling edge deployment and long-context processing without quadratic VRAM bloat.

End-to-end real-time streaming data pipeline on AWS. Ingests via Kinesis/Kafka, transforms with Apache Spark (PySpark), and lands into a scalable lakehouse — fault-tolerant, schema-enforced, built for high-throughput analytics.

Adaptive AI tutoring system using LLM-powered Socratic conversation and prerequisite-aware concept delivery. RAG backend with session-aware state management for personalized psychology education at scale.

End-to-end NLP ecosystem for emotional sentiment and audience perception analysis. Combines classical models with LLMs & RAG to deeply analyze ambiguity, sarcasm, and complex contextual framing in social media content.