Hi, I'm Emad.
I build intelligent systems.
I've developed strong expertise in Natural Language Processing, Machine Learning, Deep Learning, and Software Engineering across academic and industrial settings. I'm highly motivated to tackle challenging AI problems — from LLM-powered products to search and knowledge graph systems.
Experience
Download Resume PDFSenior Research Engineer
Thomson Reuters
Toronto, Canada
Details & Impact
Delivering software components and ML Pipelines for experimentation & deployment of AI Models (ML/DL/LLM-based models) to improve customers' flagship products, and add new AI features related to NLP, search, knowledge graph, and recommendation problems.
NLP & Machine Learning Engineer
INAGO INC.
Toronto, Canada
Details & Impact
Deep neural network development, managing datasets, training, and deployment for Language Understanding Engines and Automated Text Generation Models. [Python, PyTorch, Word Embedding, AWS EC2, SpaCy, NLTK, BERT, T5 models]. Managing collaborative research projects with universities related to Automated Text Generation and Linguistics.
Education
M.Sc. in Computer Science
York University
Toronto, Canada
GPA: 8.17 / 9
Thesis: Interactive Question Answering Using Frame-based Knowledge Representation
B.Sc. in Computer Engineering
Amirkabir University of Technology
Tehran, Iran
GPA: 17.18 / 20
Technical Skills
Projects
GenAI Summary for Business Entity Reports
2024Using LLMs prompting to generate summaries with customer-focused insights to facilitate the use of long business reports. Improving the current data extraction pipeline from reports.
Design and Building New Entity Matching System
2022 - 2024Worked on new generation of flagship product for Due Diligence and Risk Management.
New Search Improvement Feature to Promote More Relevant Documents
2021Delivered new search functionality to detect more general type of queries and questions and promoting the more suitable content for these queries for a Tax Research web application (in-production).
Improving Automated Question Generation from Documents
2020Fine-tuning T5 Transformer Language Model, experimenting for improving model input, using BLEURT evaluation. Auto-generated questions resulted in 40% reduction in manual data curation efforts.
Improved Language Understanding Engine Models
2019Training domain-specific Word2Vec word embeddings and adding detailed model testing for LSTMs to improve NLU model interpretability.
Conversational Question Answering
2018Creating a question answering dialogue system powered by syntactic and semantic analysis of documents and ontology generation. Domain-specific question answering using a dialogue interface. Part of a collaborative project with an industrial partner.
Publications
Question-worthy sentence selection for question generation
Interactive Question Answering Using Frame-based Knowledge Representation
Time aware topic based recommender system
A study on prediction of user's tendency toward purchases in websites based on behavior models