Mohammed Owais Shaikh

Junior AI/ML Developer working on Machine Learning and Deep Learning solutions using Python, PyTorch, TensorFlow, and Scikit learn. Focused on data preprocessing, model training, evaluation, optimization, and AI driven product development.

I view technical instruction at YuHasPro Nerul as the ultimate exercise in architectural mastery. Here, I bridge the critical gap between raw data science and advanced artificial intelligence by dismantling complex concepts into pure engineering logic. I lead students in building predictive models and data architectures from the ground up, utilizing Python and SQL to transform chaotic data into clean machine learning inputs. Teaching these fundamental algorithms daily acts as a powerful catalyst for my own technical evolution, directly fueling my advanced research into PyTorch, Transformer dynamics, and natural language processing. By engineering the minds that will shape tomorrow, I am simultaneously refining my own blueprint for designing high impact AI solutions.

🏆 SMART INDIA HACKATHON : GRAND FINALE During the Smart India Hackathon, I led my team to the Grand Finale where we competed among more than 44,000 teams from across India. Our mission was clear and meaningful: bridge the communication gap between speech impaired individuals and the people around them. Many individuals who cannot speak rely on sign language to express themselves, yet most people do not understand it. This creates a deep communication barrier. To address this real world problem, we designed and developed a pair of AI powered smart gloves that could interpret signs and convert them into speech. 🧤 How the System Worked → Flex sensors and an accelerometer captured every hand gesture → The ESP32 Wi Fi module transmitted the sensor data in real time → An AI model interpreted the gestures and mapped them to their meaning → The recognized message was converted into audible speech → A built in speaker played the output for clear and seamless communication 🔧 Technical Side • Flex sensors to detect finger bending • Accelerometer to identify movement and orientation • ESP32 microcontroller for real time wireless communication • Trained AI model for gesture interpretation • Integrated speaker for voice output 🎯 Why This Project Matters This was more than engineering. It was an effort to give voice to those who communicate through silence. Our goal was not to replace sign language but to respect, translate and amplify it so people could be understood without barriers.

🏆 CIIA NATIONAL COMPETITION : FINALS In the CIIA National Competition, I led my team to the Finals where we worked on a problem that affects millions of people every day: the complexity of creating legal documents. Legal paperwork protects our rights, yet the process often feels confusing, slow, and overwhelming for the common person. Our mission was to remove this barrier and make legal access simple, transparent and effortless. 🖥️ What We Built We developed a web based platform that automates the process of generating legal documents. The goal was to replace long procedures and unnecessary middlemen with clarity, speed and confidence. 🧩 How the Platform Worked → Users selected the type of legal document they needed → Chose their region to match local requirements → Filled in the essential details through a guided form → Instantly viewed a clean auto generated preview of the document → Submitted the request for lawyer review and verification → Received the final document delivered directly to their doorstep 🔧 Technical Side • Dynamic form logic based on document type • Automated preview generation using structured templates • Validation checks to prevent user errors • Request routing to authorized legal professionals • Streamlined delivery system for verified documents 🎯 Why This Project Matters Legal processes should empower people, not intimidate them. By simplifying document creation into a few guided steps, we made legal access more approachable, faster, and far more reliable. This project was not just about building a platform, but about enabling people to take control of their legal needs without confusion or fear.

I executed an intensive data science workflow focusing on raw data transformation and predictive modeling. I performed rigorous data cleaning and exploratory analysis to identify key feature correlations across diverse datasets. Utilizing Python and Scikit learn, I engineered supervised learning algorithms for regression tasks and validated model performance using statistical metrics like RMSE and R2 scores to ensure high accuracy in predictive analytics.

🏆 PROJECT DEEP BLUE : SEMIFINALS In Project Deep Blue, a national level competition organized by Mastek, I led my team to the Semifinals. Our mission addressed a very real challenge seen in workplaces everywhere: long meetings that stretch for hours, making it difficult to remember key points, decisions and action items. We wanted to create a system that does the listening for you, so people can focus on discussions instead of note taking. 🖥️ What We Built We developed a platform where users could upload recorded meeting videos. The system intelligently processed the content and delivered a clear, structured summary that highlighted what truly mattered. 🧩 How the System Worked → Users uploaded their meeting recording → The system extracted the audio from the video → Speech was converted into text using speech to text processing → A deep learning model inspired by large language model (LLM) architecture analyzed the transcript → The model generated a complete transcript, identified key takeaways and produced a concise summary This allowed users to revisit important discussions in minutes instead of hours. 🔧 Technical Side • Audio extraction and preprocessing • Automated speech to text conversion • Deep learning model for summarization and key point extraction • LLM inspired architecture to understand contextual flow • Git for version control • End to end integration for smooth user experience 🎯 Why This Project Matters Meetings hold valuable decisions, but human memory is limited. By automating transcription and summarization, we made it easier for teams to recall important points, avoid miscommunication and focus on execution. This solution turned long discussions into actionable clarity.

Graduate with CGPA 8.44 in Computer Science Engineering (AI and Machine Learning), where I worked on real-world AI projects and participated in national-level innovation competitions. I reached the Grand Finale of Smart India Hackathon among 44000 plus teams with an AI based Smart Glove for speech-impaired communication, and led my team to the semifinals of the CIIA competition by developing a legal document automation system. I also built projects like Drowsiness Detection to prevent road accidents and Speech Emotion Recognition to analyze emotions from voice. These experiences strengthened both my technical skills and my soft skills such as leadership, teamwork, problem solving, and time management.