Hi, I’m Mayank Anand 👋

🎓 I’m a Computer Science undergraduate at the Indian Institute of Information Technology, Allahabad (IIIT-A), pursuing a B.Tech in Information Technology (2024–2028).

I’m deeply interested in building intelligent systems at the intersection of Robotics, Computer Vision, and Medical AI, with a strong inclination toward optimization, spatiotemporal modeling, and decision-making under uncertainty. My work is research-driven, application-focused, and oriented toward real-world clinical impact.

Driving Thought

"Understanding motion, uncertainty, and intelligence in the physical world to build safer, smarter medical AI systems."

💡 What I’m currently focused on

• Medical Robotics & Embodied AI
• 3D Freehand Ultrasound Reconstruction
• Spatiotemporal Vision & Articulated Motion
• Uncertainty-aware perception & decision-making
• Reinforcement Learning for robotic navigation
• Multimodal reasoning for healthcare applications

🛠️ Technical Skills

Languages
Python, C++, C, JavaScript

ML & AI
PyTorch, TensorFlow, OpenCV, Scikit-learn
MuJoCo, ROS2, Gazebo
LangChain, LangGraph, LLM-based agents

Robotics & Simulation
ROS2, RViz, MuJoCo, SlicerROS2

Data & Dev
Git, Docker, Flask, MongoDB, SQL
Matplotlib, NumPy, Pandas

🔬 Research & Achievements

• 📄 QueST: Persistent Queries as Semantic Monitors for Drift Suppression in Long-Horizon Tracking Accepted at ICLR 2026 (CAO Workshop)
• 📄 HARPO: Hierarchical Agentic Reasoning for User-Aligned Conversational Recommendation Accepted at ACL Main 2026
• 📄 UltrasODM: A Dual-Stream Optical-Flow–Mamba Network for 3D Freehand Ultrasound Reconstruction Accepted at AAAI 2026 (AIMedHealth Program, PMLR)
  🔗 https://arxiv.org/abs/2512.07756v1
• 🧠 Working on articulated spatiotemporal tracking and reasoning at CIR, IIIT-A
• 🤖 Developed RL-based ultrasound navigation for medical imaging
• 🗣️ Built multimodal mental-health assessment systems with the Indian Army
• 📊 Working on agentic AI + multimodal datasets for healthcare at IIT Patna

🌱 My Research Journey

I began with physics and optimization, which naturally led me to machine learning. My early work focused on RL-based ultrasound navigation, helping a virtual probe reach clinically meaningful views.

This evolved into 3D freehand ultrasound reconstruction, where I tackled motion drift, brightness variation, and uncertainty under real clinical constraints. My current work blends:

• Vision (flow, tracking, reconstruction)
• Robotics (kinematics, trajectory prediction)
• AI (uncertainty, reasoning, decision-making)

I am especially excited about embodied medical AI, digital twins, and human-in-the-loop clinical systems that assist doctors rather than replace them.

🤝 What I’m looking for

I’m open to research collaborations, internships, and projects in:

• Medical AI
• Embodied Robotics
• Spatiotemporal Vision
• Surgical AI & Digital Twins

📬 Reach me

📧 Email: anandmayank698@gmail.com
💼 LinkedIn: https://www.linkedin.com/in/mayank-anand-480741231
💻 GitHub: https://github.com/AnandMayank
💬 Discord: mayanka9376