Computer scientist in AI & ML, from research to working systems.
PhD, Computer Science · University at Buffalo (SUNY)Montréal, Canada
I design and build AI systems end to end: from research and model development to fast C/C++ inference, cloud training pipelines, and prototype SaaS-style tools backed by models for batch and inference workloads. I’m as comfortable making a model faster and tracking down a bug as I am shaping the product around it.
Agentic systems and RAG pipelines on ChromaDB, plus R&D on new model architectures.
Feed-forward nets, RNNs and LSTMs implemented from scratch for real-time, low-latency use.
Training, evaluation and deployment pipelines that carry models from experiment to working use.
Audio enhancement and real-time code, plus prototyped SaaS-style internal tools running batch jobs and inference.
AI AgentsRAGModel EvaluationML SystemsReal-Time InferenceAudio AICloud & MLOpsC++ / Python
LanguagesEnglish fluentFrench functionalPersian fluent
Research · Product design · Implementation · Optimization · Debugging

Built an end-to-end RAG pipeline from the official Discover Canada citizenship guide, combining layout-aware PDF processing, semantic chunking, MiniLM embeddings, and ChromaDB retrieval. Evaluated retrieval quality and built an interactive AI-powered study application with quizzes, flashcards, concept maps, and timelines.
Live demo · Click to try→
Built an interactive atlas of 29 years of DAFx proceedings (1,706 papers). Embedded the papers, proposed research-area and method labels, and made the archive explorable through a turning timeline, measured questions, a topic map, a methods matrix, and a coauthor network, all linking back to the official papers.
Live demo · Click to explore→