• Designed a Model Context Protocol (MCP) Generation System that converts APIs into MCP servers; achieved 95% faster API integration, 10x faster deployments, and 100% security compliance

• Developed a Swagger Generation pipeline (10,000+ lines of production JS) that parses repositories, auto-documents endpoints with LLM-assisted annotations, and supports OpenAPI 2 to 3 conversion; eliminated manual Swagger creation for customers, significantly reducing onboarding time.

• Integrated Swagger/MCP generation into both the CodeGlide.ai web platform and GitHub Actions workflow

• Enhanced enterprise adoption with endpoint selection tools, schema validation, and automated error handling.

• Developed an Adaptive LLM Routing System that dynamically selects between multiple foundation models (OpenAI, Anthropic) using semantic similarity, prompt classification, and historical metrics, reducing cost by 40%

• Researcher from January 2023 - June 2024 in Professor Razvan Marinescu's Research Group

• Initial research into Diffusion Models for MRI generation

• Developed a Diffusion Model from scratch using PyTorch, Transformers, numpy, and Convolutional Neural Networks.

• Implemented key components including UNet, encoder, decoder, upsampling, downsampling, and self-attention mechanisms

• Explored Audio Machine Learning for healthcare applications, focusing on signal processing and machine learning techniques

• Developed an Ensemble Learning Model combining a multi-feature CNN and XGBoost for respiratory disease classification

• Applied advanced audio signal processing techniques such as Mel Spectrograms, Chroma Short Time Fourier Transforms, Mel Frequency Ceptral Coefficients, Spectral Rolloff, Spectral Centroid, and Zero Crossing Rate

• Implemented various data augmentation methods, including time-stretching, pitch-shifting, and noise addition

• Authored a research paper available on arXiv, currently awaiting journal publication as I actively submit to additional conferences.

• Coordinated communications between Santa Cruz Artificial Intelligence (SCAI) and the university to ensure effective collaboration and support for club activities.

• Designed and developed an inclusive curriculum accessible to all students by incorporating foundational AI concepts and advanced topics to cater to diverse skill levels.

• Presented the curriculum and guided members through projects during meetings to achieve a hands-on learning environment and encourage active participation.

• Networked with industry professionals to secure guest speakers to provide members with valuable insights and exposure to real-world AI applications.

• Restructured the leadership team and club strategy to better incorporate new members and students

• Developed and implemented new marketing campaigns to attract and retain members, thus significantly increasing club visibility and engagement.

• Collaborated with leadership on various responsibilities, including curriculum design, marketing, financial management, project coordination, and outreach initiatives.

• Helped organize AI competitions, providing members with opportunities to apply their skills in a competitive and collaborative setting.

• Hosted talks for other clubs on campus to promote interdisciplinary collaboration in AI.

• Networked with corporations to explore potential club sponsorships to secure resources and support to enhance club activities and member benefits.

• Achieved a 500x growth in club membership under my presidency, which ultimately finished with SCAI as being one of the largest organizations on campus.

• Led instructional meetings, teaching members on various AI-related topics and guiding them through hands-on projects to enhance their practical skills.

• Developed and refined the curriculum to keep the material fresh for new members.

• Oversaw member projects, providing technical guidance to ensure members successfully complete their projects

• Facilitated effective communication between Santa Cruz Artificial Intelligence (SCAI) and the University regarding club initiatives and promotions

• Gained an understanding of basic Machine Learning fundamentals, including key concepts such as supervised and unsupervised learning, model evaluation, and common algorithms like linear/logistic regression, clustering, and eventually dove into more complex Neural Networks.

• Developed skills in reading and analyzing research papers to inform project design and implementation strategies.

• Completed various mini-projects, applying machine learning techniques to real-world problems and gaining hands-on experience in data preprocessing, model building, and result interpretation.

• Collaborated with other members on group projects, most notably working on an LSTM (Long Short Term Memory) solution for identifying intron and exon sequences in DNA.