Hello, I'm Shihab šš»
With hands-on internship experience and as a Computer Science student, Iām dedicated to developing software solutions that bridge technology with real-world needs. My expertise spans AI, backend, frontend, and full-stack development, enabling me to contribute to projects ranging from early-stage prototypes to production-ready systems. I focus on writing clean, maintainable code and building user-centric, engineering scalable solutions while thriving in environments that prioritize innovation and continuous learning.
Feel free to explore my projects below or connect with me on GitHub and LinkedIn. Iām always open to discussing new ideas, collaboration, or internship opportunities.
Projects
See all projects-
AgentPilot
AgentPilot turns natural-language instructions into browser actions, such as opening an allowed website, filling a form, or checking page content. An AI agent uses browser tools exposed through the Model Context Protocol (MCP), inspects page snapshots, and verifies the result. Configure with Siri via Shortcuts to automate daily browsing sessions.
TypeScript MCP OpenAI Playwright Cloudflare Python FastMCP -
Emotion Classification ML vs AI
A simple emotion classification system with two flask API endpoints, one is SVM machine learning model and other is openai gpt model. This system classifies emotions based on 6 categories (e.g. 0: sadness, 1: joy, 5: surprise).
Python Flask Scikit-learn OpenAI -
AI Chatbot
Built a full-stack open-source chatbot with chat history, role persistence, image generation, and secure API key handling. Designed a responsive UI and clear docs and contribution guidelines, attracting contributors an enabling smooth community contributions.
Spring Boot React Docker Render
Work Experience
FGF Brands
AI Solutions Engineer (Co-op)
- Developing full-stack features for FGF's internal AI agent platform, using Python, FastAPI, FastMCP, React, TypeScript, Azure, and PostgreSQL.
- Building REST APIs and Model Context Protocol (MCP) integrations that enable AI agents to invoke enterprise tools, execute workflows, and interact with external services.
- Integrating enterprise connectors to automate business operations, including sending emails, posting Microsoft Teams messages, and generating business dashboards.
- Implementing backend services, database queries, and application logic using Azure PostgreSQL, including secure authentication and Microsoft 365 integrations through MSAL.
- Designing reusable frontend and backend components for agent configuration, tool execution, and conversational user interactions.
- Collaborating with software engineers using Jira to develop, test, debug, and deploy application features in an Azure-hosted production environment.
Trillium Health Partners
Information Systems (Co-op)
- Developed and deployed Python, PowerShell, and Power Automate scripts to automate monthly security patching workflows across 50 clinical applications.
- Automated repetitive patching tasks, saving an estimated 16 hours of manual work per month.
- Implemented error handling, logging, and end-to-end testing to improve the reliability of long-running automation processes.
- Automated the collection of application validation results, screenshots, and execution errors to support patching documentation and reporting.
- Collaborated with application analysts, IT teams, and external vendors to translate operational requirements into maintainable automation solutions.
- Deployed and documented attended and unattended automation workflows within Windows server environments.
AI for Impact (Club)
Machine Learning Engineer
- Collaborated with a student-led research team at York University to explore machine learning and computer vision applications addressing healthcare challenges.
- Evaluated open-source machine learning models from Hugging Face for brain tumor detection and compared model performance for experimental research.
- Investigated practical applications of computer vision and deep learning models in healthcare-focused research.
- Developed and delivered technical workshops introducing Generative AI, AI agents, and real-world AI applications to York University students.
Vosyn
Backend Developer Intern
- Designed and developed Python/Django microservices and REST APIs for VosynVerse, an AI-powered content platform supporting real-time translation and transcription.
- Engineered and documented backend features supporting an AI-powered smart video summarization widget.
- Designed PostgreSQL database schemas and optimized SQL queries to improve backend data retrieval performance.
- Built and deployed containerized backend services using Docker, Terraform, GitHub Actions CI/CD, and Google Cloud Platform (GCP).
- Leveraged GCP and Firebase services to optimize cloud storage and application data retrieval.
- Applied software development lifecycle (SDLC) and test-driven development practices, writing unit tests with pytest and validating REST API responses using Postman.
- Collaborated with frontend developers, QA engineers, and product team members in Agile sprints to deliver end-to-end application features.