Available to hire
Hi, I’m Paria Zabihitari, a passionate Machine Learning Engineer based in Toronto with over two years of experience in designing scalable ETL/ELT pipelines and building predictive machine learning models. My work ranges from developing AI-powered chatbots to implementing advanced NLP techniques, combining my technical skills with data-driven decision-making to solve real-world problems.
I thrive on creating innovative solutions using frameworks like PyTorch and TensorFlow and enjoy collaborating closely with industry stakeholders to ensure impactful outcomes. I’m always eager to learn and leverage AI technologies to enhance business processes and user experiences.
Skills
Experience Level
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Intermediate
Intermediate
Intermediate
Intermediate
Intermediate
Intermediate
Intermediate
Language
English
Fluent
Persian
Fluent
Work Experience
Data Engineer Intern at Leni
May 31, 2025 - July 19, 2025Led development and enhancement of an AI-powered chatbot using LangChain, Vertex AI, and LLM APIs to enable property owners to retrieve and manage data via natural language queries. Integrated RAG pipelines with Pinecone and FAISS for scalable semantic search, increasing document retrieval accuracy by 10%. Deployed and evaluated the chatbot using Postman for testing and monitored production performance via AWS CloudWatch. Debugged issues by analyzing logs and user interactions, leading to improved chatbot accuracy and user satisfaction. Employed advanced prompt refinement and integrated Cube’s semantic layer, boosting coverage by 36%, accuracy by 25%, and reducing latency by 10 seconds, resulting in a 10% increase in user satisfaction.
Research Assistant at Toronto Metropolitan University
December 31, 2024 - July 19, 2025Designed graph-based reasoning systems using Graph Neural Networks on TigerGraph to enable structured knowledge retrieval for Unilever’s product and customer data. Curated and structured GSQL training data to fine-tune LLMs for graph query generation, automating complex retrievals across large-scale graph databases. Applied Chain-of-Thought prompting, RAG, and Graph-RAG techniques to reduce hallucination and improve retrieval accuracy in LLM outputs. Co-authored a peer-reviewed paper published at IEEE SysCon 2025 on hallucination mitigation in graph-augmented LLM systems.
AI Engineer Intern at WebsiteToon Digital
February 29, 2024 - July 19, 2025Designed and implemented end-to-end pipelines to preprocess unstructured financial statements for accurate extraction of key financial ratios and structured data for analysis. Applied advanced prompt engineering, Retrieval-Augmented Generation (RAG), and document chunking to enhance information retrieval precision and relevance. Delivered clear, stakeholder-focused summaries and insights by combining post-processing techniques with financial context to support data-driven decision-making for investors.
Data Scientist Intern at Flybits Co.
August 31, 2023 - July 19, 2025Transformed raw data into clean, structured datasets for machine learning by implementing advanced outlier detection, imputation techniques, domain-specific feature engineering, normalization, class balancing, and dimensionality reduction to enable robust credit risk modeling. Conducted exploratory data analysis and visualized SME behavior using clustering methods to uncover credit risk patterns and guide financial segmentation. Created presentation-ready visualizations and dashboards using Seaborn, Matplotlib, Tableau, and Excel. Designed and implemented time series forecasting models leveraging engineered temporal features with LSTM and ARIMA, achieving 95% prediction accuracy for SME cash flow forecasts. Collaborated with industry stakeholders such as TD Bank and CEVA Logistics to communicate analytical findings and ensure project alignment with business goals.
Research Assistant at Toronto Metropolitan University
December 31, 2024 - July 22, 2025Designed graph-based reasoning systems using Graph Neural Networks on TigerGraph for Unilever's product and customer data. Curated and structured GSQL training data to fine-tune LLMs for graph query generation, automating complex retrieval across large-scale graph databases. Applied Chain-of-Thought prompting, RAG, and Graph-RAG techniques to reduce hallucination and improve retrieval accuracy. Co-authored a peer-reviewed paper published at IEEE SysCon 2025 on hallucination mitigation in graph-augmented LLM systems.
Data Science Intern at Myant Corporation
August 31, 2024 - July 22, 2025Developed an end-to-end time series analytics pipeline for ECG data including temporal feature engineering, time-aware imputation, and longitudinal preprocessing. Uncovered dominant heart rate patterns across cycles using unsupervised clustering to support behavioral segmentation. Modeled baseline physiological trends using LSTM autoencoders, leveraging reconstruction error to detect significant deviations with 94% precision.
Education
MEng. Computer Engineering with AI Concentration at Toronto Metropolitan University
September 1, 2022 - February 28, 2024B.Sc. Electrical Engineering at Shahid Beheshti University of Iran
September 1, 2016 - September 30, 2020MEng at Toronto Metropolitan University
September 1, 2022 - February 28, 2024B.Sc. at Shahid Beheshti University of Iran
September 1, 2016 - September 30, 2020MEng at Toronto Metropolitan University
September 1, 2022 - February 28, 2024Qualifications
Industry Experience
Software & Internet, Financial Services, Education, Real Estate & Construction, Professional Services, Healthcare, Transportation & Logistics, Consumer Goods
Skills
Experience Level
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Intermediate
Intermediate
Intermediate
Intermediate
Intermediate
Intermediate
Intermediate
Hire a AI Engineer
We have the best ai engineer experts on Twine. Hire a ai engineer in Toronto today.