Inspiration

The genesis of this project lies in my personal experiences as a first-year college student abroad during the COVID-19 pandemic. Confronting numerous challenges and pressures, I recognized the profound need for a virtual assistant that could offer immediate, accessible, and tailored educational support. While the university provided mentors for every 50 students, their availability was limited, especially during off-hours, leading to delayed responses and support. This experience fueled my commitment to developing a solution that could offer real-time, personalized guidance to students in need.

Objectives

• Personalized Learning Assistance: Offering one-on-one support to students in mastering complex subjects, particularly in database management and object-oriented programming.
• 24/7 Availability: Ensuring students have access to assistance at any hour, especially during late-night study sessions or when human mentors are unavailable.
• Emotional Support and Engagement: Utilizing empathetic and engaging interactions to keep students motivated and confident in their learning journey.

Conclusion

The AI-Enhanced Learning Companion is more than just a chatbot; it's a testament to the power of technology in transforming educational experiences. By providing personalized, empathetic, and continuous support, it stands as a beacon of innovation in the realm of educational technology, significantly contributing to the betterment of students' academic and personal growth.

Researched and implemented a deep learning model designed to improve the performance of Gender and Age Recognition Models when applied to individuals wearing medical masks.

Using advanced techniques in unsupervised domain adaptation, Developed a PyTorch‑based deep learning model, Achieved a notable 5% increase in gender recognition accuracy on masked faces compared to the state‑of‑the‑art models, demonstrating the effectiveness of the innovative solution.

In this project, I was mainly responsible for the literature review, evaluating the performance of the state-of-the-art models, and validating the robustness of the developed models.

The "leFt BehInd" project at the Interactive Media Lab offers a unique exploration of Aurora's neighborhood using interactive media, digital storytelling, and immersive technology.

It features a panoramic binocular setup facing a projection of Corso Novara in Aurora District of Turin. Users can interact with the projection by directing the binoculars towards highlighted areas, triggering access to audiovisual content. This creates an engaging and interactive experience.

The project is powered by Isadora, a real-time data processing software essential for animating digital materials like video, sound, and light. Isadora enables real-time manipulation of input values, facilitating a variety of design responses to diverse data sources. The project's main goal is to provide a personal perspective on the life in Aurora, using binoculars as a physical interface to symbolize an outsider's view of the neighborhood.

This research project centers on the critical role of AI in reshaping the dynamics of the modern workplace, with a particular emphasis on the rise of remote work, a trend accelerated by the COVID-19 pandemic.

It investigates whether telework will remain a staple in future work environments and examines its economic, environmental, and productivity implications. By analyzing specific cases in Italy and Denmark, the project underscores the influence of cultural and societal factors, addressing key challenges such as gender disparities, privacy issues, and maintaining work-life balance in an AI-driven work environment. It proposes various solutions to address these challenges, including gender-specific policies and transparent workplace monitoring.

The goal is to foster a future where enhanced awareness, thoughtful policies, and empathetic AI technologies amalgamate to create a balanced and effective work setting. This initiative underscores the need for AI literacy and promotes a diverse and inclusive AI workforce, emphasizing the importance of intellect, creativity, empathy, and appropriate skills in an AI-driven era.

In this project, it is tried to train a chatbot using RASA to recommend food recipes. RASA is an open-source machine learning framework for automated text and voice-based conversations. This bot is called “Be My Chef” or in short form “BMC”.

The BMC is designed to recommend recipes tailored to dietary allergies and drug interactions. This project specifically targets elderly users who find traditional technologies challenging, offering voice-activated assistants like Alexa. I aim for a healthier living, AI literacy, and ethical AI use, contributing towards a more equitable world, particularly in the context of aging and technology.

This chatbot uses the user's input to determine what cuisine the user is looking for and what ingredients they want their meal to have. The user can also specify the ingredients that they do not like or are allergic to. The BMC provides the user with a couple of recipes by telling them the name of the food and the time that is required for the meal to be cooked. The user can add or remove ingredients when the BMC recommends food with a specific ingredient that they do not like, and they want to prevent the bot from recommending it again. The BMC moves through the recipes until the user affirms that they like it. At the end, when the user likes a recipe, the BMC sends an SMS to their phone which contains the name of the recipe and a URL of the detailed recipe description.

The chatbot created using RASA has mainly two models; The core and the NLU model. The NLU model is trained on a set of naturally spoken sentences annotated by the entities that are included and what the user may intend by saying them. The core model, on the other hand, is trained on stories; A set of possible paths to talk to the chatbot.

Alternate Reality Game (ARG) "L'Ultimo Capitolo" presents an immersive and participatory experience, blending a compelling narrative with interactive elements. The sroty revolves around the disappearance of Stefano Rovetti and the subsequent investigation led by his friend and protagonist, Giulia Olivetti.

Utilizing a multi-platform approach, the narrative unfolds through Instagram, Telegram, and an enigmatic website called the "Libro delle Femmine." Participants, interacting with the story's elements, decipher enigmas, and collaborate in a virtual community, slowly unraveling the extreme ideals of the "club del libro" and its sinister plans. The immersive experience engages users in real-time challenges, culminating in the discovery of a planned terrorist attack. The players must race against time to prevent the catastrophe, showcasing the ARG's emphasis on collaboration, critical thinking, and the exploration of social issues related to extremism and justice.

The story, rooted in the high school setting of "Galileo Ferraris" in Turin, delves into the aftermath of a feminist group inspired by the novel "Memorie di una ragazza interrotta." The group, known as the "club del libro," becomes entangled in the mysterious disappearance of Stefano Rovetti. Giulia Olivetti, suspecting the group's involvement, enlists online participants to infiltrate the club, leading to the discovery of a hidden agenda involving a planned bomb attack. The ARG weaves together themes of extremism, justice, and the dangers of unquestioning adherence to radical ideologies, providing users with an interactive and thought-provoking experience that encourages collaboration and critical reflection on societal issues.

The InSitu League, aimed at under-16 students and introduced for the first time in the IranOpen competitions, stands out for prioritizing equality and creativity. InSitu offers a unique approach to robotics competitions by placing participants in scenarios that emphasize problem-solving skills.

Contrasting with traditional leagues, where competitors bring pre-assembled robots, InSitu encourages participants to showcase their robot-building abilities on-site. Upon arrival at the competition venue, participants are provided with a designated workspace resembling a basic workshop. The competition takes an innovative turn as the committee presents challenges for the participants to solve.

To address these challenges, a unique robotics accessories market is established on-site, allowing participants to purchase necessary components using a special currency provided to each team at the beginning of the competition. It is worth mentioning that students were not allowed to use any components from outside the market. This approach not only fosters financial management skills but also ensures an equitable playing field for all competitors, irrespective of their socio-economic backgrounds.

The theme of the competition was related to major global challenges and the preservation of endangered animals. During the 3-day event, students learned about robotics, the effects of climate change, and the new technologies of the future.

As a Technical Committee member of this league, I planned each day's challenges to be both stimulating and manageable for students with an elementary level of education. We assisted them during the event by holding educational workshops and providing technical support throughout the competition.

Pishrobot enhances knowledge and skills through game‑based education in Robotics and STEM for kids, employing Scratch coding and the Pbot robot. Programming for the physical robot provides a deeper and more tangible experience compared to simulated environments, allowing children to gain valuable hands‑on experience in real‑world challenges.

• Utilized Scratch coding and Pbot (a specialized robot) to facilitate interactive learning experiences, enabling kids to acquire pro‑ gramming skills through engaging activities.
• Developed and delivered a systematic educational curriculum, fostering a deep understanding of coding by guiding students through real‑world challenges and hands‑on projects.

Developed two autonomous soccer player robots in a team of 5 using image processing for ball detection and implemented efficient algorithms to create a functional system with high accuracy.

• Used Arduino, AVR, Altium Designer, and OpenCV
• Selected as the country's representative in RoboCup Canada 2018
• Achieved second place in the IranOpen competition in 2016 and 2018
• Named the best technical team of 2018 competitions for exceptional HW/SW design