About me
I build robotics, UAV and industrial-automation systems — and the software that runs them
— from computer vision to large-scale machine learning systems.
Experience
Building the backbone of modern AI—delivering production-grade systems with mathematical rigor and operational excellence
- 2020 — Present
Founder & CTO [ Orbitronix Technologies ]
Building enterprise software, agritech automation (FarmOS Pulse) and robotics systems for clients and institutions.
- 2022
Aerospace Researcher [ IAC 2022, Paris ]
Presented research on modelling and validating the Moon's radiation environment at the 73rd International Astronautical Congress.
- Ongoing
STEM Educator [ Space Camp India ]
Runs robotics and space-science workshops, and mentors UAV and autonomous-swarm teams.
My Process
I integrate deep architectural research, rigorous data strategy, and production engineering to build resilient AI systems.
01. System Audit & Discovery
I begin with an in-depth audit of your data landscape, current infrastructure, and core business objectives. This foundational phase identifies technical constraints and sets the architectural direction for the project.
02. Architectural Strategy
Together, we develop a comprehensive technical roadmap. I design the neural architecture and data flow, establishing clear performance benchmarks—such as latency thresholds and accuracy targets—required for success.
03. Engineering & Deployment
The development phase moves through focused sprints of training, fine-tuning, and rigorous testing. I transform theoretical designs into scalable, production-grade AI models integrated into your live environment.
04. MLOps & Evolution
Post-deployment, I implement continuous monitoring and MLOps pipelines to prevent model drift. We constantly measure and refine the system, ensuring the AI remains accurate and scalable as your data demands evolve.
Tech Stack / Tools
I fuse scalable AI architecture, data-driven strategy, and real-world deployment expertise to build reliable intelligent systems.
Languages
Python
C++
JavaScriptFrameworks
PyTorch
TensorFlow
Scikit-learn
Data
Pandas
NumPy
SparkMLOps
Docker
Kubernetes
MLflow
Cloud
AWS
GCP
AzureFrequently Asked Questions
Your questions about my process, services,
and workflow—answered.
Robotics, UAV and industrial-automation systems, and the software that runs them - from embedded firmware and sensor fusion through to production web platforms.
With an audit of the existing hardware, data and infrastructure. Constraints get mapped and the architecture agreed before any code is written.
Bangladesh, working remotely with clients in Europe and North America - taking projects from the first architecture decision through to a running production deployment.
You do. Codebase, infrastructure and documentation transfer in full at the end of the engagement. No vendor lock-in.