AI/ML Engineer | Gen AI Specialist | Building tools for real-world impact
🎓 MS in Artificial Intelligence, Khoury College of Computer Sciences, Northeastern University
Passionate about building intelligent systems that solve real-world problems, with expertise in computer vision and generative AI.
I turn complex AI concepts into practical applications and enjoy teaching computers to see the world (while occasionally apologizing to my GPU).
- 💡 Currently developing GrantWell at The Burnes Center for Social Change, helping underserved communities access federal grants
- 🧠 Published researcher in healthcare AI, focusing on radiation safety and personalized dosimetry in CT imaging
- 🤖 Experience with generative AI, RAG techniques, and deploying AI solutions at scale
- 🌐 Built end-to-end MLOps pipelines and production-ready AI systems using AWS and GCP
I'm passionate about AI for social impact and creating technology that makes a meaningful difference.
Languages
Python • Java • C++ • TypeScript • React • HTML/CSS
Machine Learning & Data Science
TensorFlow • PyTorch • OpenCV • HuggingFace • Transformers • NLTK • MLFlow
Cloud & MLOps
AWS (Lambda, Bedrock, DynamoDB) • GCP (Vertex AI) • Docker • Git
AI-powered platform helping underserved communities in Massachusetts pursue federal grants.
Tech: React • AWS (Lambda, Bedrock, DynamoDB) • RAG
- Co-developed a public-facing AI tool with the Massachusetts Federal Funds Office
- Designed an AI-powered grant recommendation system using Claude 4 via Amazon Bedrock
- Built a guided grant writer interface with real-time AI assistance and document processing
- Conducted user demos across Massachusetts to gather feedback and improve features
Real-time music recommender using facial emotion analysis and mood-based clustering.
Tech: Python • Docker • Airflow • DVC • MLflow • Vertex AI
- Built a system that detects user emotions via facial analysis and maps them to song clusters
- Designed a complete MLOps pipeline with Docker, Airflow, and DVC
- Deployed models using Vertex AI and tracked performance with MLflow
- Optimized data workflows with parallel processing and CI/CD implementation
Transformer-based sign language recognition system for Indian Sign Language.
Tech: PyTorch • Transformers • Computer Vision
- Built system using heatmap-guided pretraining and binary masking
- Improved recognition accuracy by 85%
- Supported future deployment for UMANG services as part of C20 India 2023 Digital Education initiative
Jun 2024 - Aug 2024
- Conducted research on AI applications in radiation safety and personalized dosimetry in CT imaging
- Co-authored two peer-reviewed journal articles on AI-driven methods for medical imaging
- Synthesized findings from over 100 academic studies on CNN-based segmentation and deep learning denoising
- Peer Reviewer, European Journal of Radiology Artificial Intelligence, 2025
- "Enhancing Radiation Safety in CT Imaging through Artificial Intelligence," Physica Medica – European Journal of Medical Physics, 2025 (under review)
- "Real-time patient-specific-dose in CT through use of artificial intelligence," Journal of Radiological Protection, 2024
I'm always interested in collaborating on AI research, MLOps, and computer vision projects.
- 📧 Email: chathankandy.a@northeastern.edu
- 💼 LinkedIn: linkedin.com/in/anjithprakash
- 💻 GitHub: github.com/Anuttan
- 🏙️ Location: Boston, MA
"Teaching computers to see the world and occasionally apologizing to my GPU."