Curriculum Vitae

AI Software Engineer

Aivida Inc.

Details of my work are protected under a non-disclosure agreement (NDA).

Also see our main product: ScribbleHealth.com.

• Contributed to Scribble, an AI-powered medical documentation software, by building RESTful APIs with Django, Django REST framework, and Celery.
• Performed AI engineering tasks and utilized LLM integration frameworks like LangChain, optimizing AI components by achieving a 67% AI inference cost reduction, a 57.15% AI inference speed boost (35 to 15 seconds), and eliminating hallucinations.
• Took full ownership of designing a medical chatbot using Large Language Models (LLMs) that supports various functions/tools to handle dynamic queries, including medical note modification and medical letter generation based on patient-doctor encounters.
• Optimized the Automated Speech Recognition (ASR) system by developing a new real-time AI-powered ASR solution, reducing the cost of ASR to almost one quarter (1/4) of the original version (75% reduction).
• Developed backend and AI-based unit/integration test cases to enhance Scribble’s test coverage using PyTest and DeepEval.
• Deployed AI models using leading cloud infrastructure to optimize performance and reduce latency in AI inferences.
• Built an evaluation and data-gathering pipeline for Large Language Models (LLMs) and Automated Speech Recognition (ASR) systems, enabling meaningful comparisons between different providers and services.

Backend Engineer

Purple Stones

• Developed an E-Commerce platform with Python, Django, Celery, and PostgreSQL, creating RESTful APIs, integrating and extending Sharetribe’s features, and connecting to Stripe for payment processing.
• Collaborated on integrating Webflow UI with Sharetribe APIs to enhance website functionality by using JavaScript.
• Managed Scrum artifacts such as the Product Backlog and Sprint Backlog, employing Agile best practices for prioritization and refinement.
• Deployed the app using Docker, Nginx, and Gunicorn to support high availability, performance, and scalability in a production environment.

Machine Learning Researcher

York University

• Led 2 research projects and contributed to more than 8 other research projects in the Machine Learning for Software Engineering (ML4SE) field and published 5 papers in reputable venues such as TOSEM, AIware, and SIGIR.
• Explored the applications of Large Language Models (LLM) through Prompt Engineering in Software Engineering fields, such as Static Analysis, Program Repair, Program Slicing, and Assurance Cases.
• Utilized LLM integration frameworks like LangChain to efficiently work with various LLMs from different providers (OpenAI's GPT, Anthropic's Claude, Meta's Llama, Mistral AI, Amazon's Titan, etc.).
• Discovered 6 new real-world bugs using Fuzzing techniques in prominent repositories like TensorFlow and PyTorch, later confirmed and fixed by their respective development teams, and 2 of them have been published by the National Vulnerability Database (NVD).
• Investigated fairness of ML-based code reviewer recommender systems for the first time, reducing existing biases by 24%.
• Fine-tuned LLMs like CodeT5 and CodeBERT using PyTorch and TensorFlow for automated program repair of Java bugs.

Full-Stack Engineer, Co-founder

Papyrus

I co-founded a startup called Papyrus, which was a social media platform designed for book lovers. It allowed users to tweet, comment, and rate books, join reading groups, and share their thoughts and reading lists with others.

• Designed a social media platform for book lovers, named Papyru, by leveraging React.js, Redux, Next.js, Tailwind, Python, and Django.
• Applied Test-Driven Development (TDD) to enhance test coverage, leveraging Jest and PyTest libraries.
• Utilized Webpack for efficient code asset bundling and TypeScript to maintain type safety and detect errors throughout development.
• Managed various responsibilities across non-technical domains, including business operations, compliance, and marketing

Software Engineer

Isfahan University of Technology

• Developed a dormitory management platform and dashboard using Python, Django, PostgreSQL, JavaScript, JQuery, and Bootstrap to manage over 500 residents across the university’s dormitories and housing departments.
• Led a team of three developers, coordinating their efforts to ensure successful delivery and integration of high-quality codes.

Frontend Engineer

Mizangostar Corporation

Mizangostar, a consultancy company operating from Isfahan Science and Technology Town (ISTT), engaged me in a project aimed at developing a management dashboard. This dashboard was designed to streamline the logistics company’s operations, allowing them to effectively manage both their trucks and drivers. -- Link to the developed dashboard

• Implemented a management dashboard using JavaScript, React.js, and Material UI, leading to a notable 20% enhancement in the application’s performance compared to the previous version.
• Integrated the frontend application with the company’s logistic backend server to handle more than 1000 trucks and drivers.

Webmaster

IEEE YorkU Student Branch

In charge of the development and maintenance of the website and the ongoing resolution of IT-related matters.

• Planned and organized social events for the IEEE community, such as Industry Night and Scientific Workshops.

Mentor

Isfahan University of Technology's Web Development Chapter

• Crafted a web developer roadmap tailored for incoming freshmen, assigning and managing sample projects and tasks that empowered over 80 students to acquire essential web development skills.
• Organized web development and career-related seminars and social events for freshmen.

Master of Computer Engineering, MASc.

York University

Sofware Engineering, Machine Learning for Software Engineering (ML4SE), AI Fairness, LLM4SE

• Data Mining
• Mining Software Engineering Data
• Performance Modeling of Computing Systems

Bachelor of Computer Engineering, BSc.

Isfahan University of Technology

Sofware Engineering

• Basic Programming (C)
• Advanced Programming (C++)
• Data Structures
• Algorithm Design
• Database Design
• System Analysis and Design (Software Engineering I)
• Agile Development and Scrum (Software Engineering II)
• Signal and Systems
• Machine Learning and Computational Intelligence
• Artificial Intelligence
• Business Intelligence and Data Warehouse

York University Graduate Fellowship

York University

$62,500 for my master's study, including TAship and RAship funding.

Member of Iran National Organization for Development of Exceptional Talents (NODET)

Iran National Organization for Development of Exceptional Talents (NODET)

Using GPT-4 Turbo to Automatically Identify Defeaters in Assurance Cases

2024 IEEE 32nd International Requirements Engineering Conference Workshops (REW)

History-Driven Fuzzing For Deep Learning Libraries

ACM Transactions on Software Engineering and Methodology (TOSEM 2024)

Effectiveness of ChatGPT for Static Analysis: How Far Are We?

In Proceedings of the 1st ACM International Conference on AI-powered Software (AIware 2024)

Assessing the Impact of GPT-4 Turbo in Generating Defeaters for Assurance Cases

FORGE '24: Proceedings of the 2024 IEEE/ACM First International Conference on AI Foundation Models and Software Engineering

Fairness Analysis of Machine Learning-Based Code Reviewer Recommendation

5th International Workshop on Algorithmic Bias in Search and Recommendation (Bias@SIGIR2024)