I enjoy building intelligent systems that combine strong mathematical foundations with practical engineering workflows — ranging from deep learning experimentation and reproducible ML pipelines to full-stack AI applications and deployment-oriented systems.
class Biswarup:
def __init__(self):
self.role = "Tech Nerd"
self.education = "M.Sc. Data Science"
self.interests = [
"Agentic Ai Systems",
"Deep Learning", "Spatiotemporal Learning",
"Statistical Modeling", "ML Pipelines",
"Research & Industry-Oriented AI Systems"
]
self.philosophy = "Strong models survive reality, not just benchmarks."
def current_focus(self):
return "Building reproducible, mathematically grounded AI systems"- 🧠 Deep Learning & Neural Networks
- 📈 AI Model Development & Evaluation
- 🌐 Spatiotemporal Learning
- 🔁 Machine Learning Pipelines
- 📐 Statistical Thinking in AI
- 🧪 Reproducible ML & Experimentation
- 🎓 Research & Industry-Oriented AI Systems
|
🧠 AI / Machine Learning 
📊 Data & Analytics 
|
🚀 Engineering & MLOps 
🌐 Development 
|
|
♟️ Chess AI with Basic Optimizer AI-driven chess engine focused on algorithmic decision-making, move evaluation, and optimization workflows. |
🌐 Full-Stack Portfolio Website Responsive portfolio platform with frontend-backend integration and dynamic content workflows. |
🧠 Deep Learning & Research Work Segmentation architectures, sequential modeling, and research-driven experimentation in PyTorch. |
- 📄 When Models Lie: The Silent Assumptions Behind Accuracy in Data Science
- 🏥 ISMRM Indian Chapter Conference Abstracts (2026)
💼 LinkedIn • 🌐 Portfolio • 📫 majumdarb104@gmail.com
"Strong models are not defined only by performance, but by how well their assumptions survive reality."