Available for collaboration · Daejeon, KR

Asif PatankarAI for satellites & drones.

AI Research Engineer turning Earth-observation imagery and aerial sensor data into intelligence — and running it in real time on the edge.

4+
years in ML & deep learning
EO · CV
satellite & aerial vision
Jetson
edge AI deployment
Type 4
licensed multicopter pilot (KR)

01 / Capabilities

From orbit to edge — the full sensing-to-decision stack.

I develop intelligent AI solutions that transform industries through research and engineering — computer vision, 3D reconstruction and embedded AI optimisation, built for the constraints of real aircraft and real imagery. Full story →

CAP.01 — Model development

Geospatial & vision AI

Advanced machine learning and deep learning for computer vision, 3D reconstruction and satellite imagery — from CNN architectures to network optimisation across aerospace, healthcare and autonomous systems.

CAP.02 — Embedded AI

Drone edge inference

Optimised AI for resource-constrained hardware with TensorRT and NVIDIA Jetson — real-time inference, CUDA programming and efficient pipelines for drones, vehicles and IoT.

CAP.03 — Vision systems

End-to-end visual intelligence

Satellite image analysis, 3D point clouds, video deblurring, object detection, RGB-D analysis, SLAM and custom algorithms — from agriculture to urban change detection.

02 / Mission log

Research that flies.

Selected projects across drone autonomy, edge inference and Earth observation. All portfolio →

MSN-01Drone · Edge AI

Embedded AI human detection

Real-time drone-based human detection system

An autonomous drone system with on-board human detection running on NVIDIA Jetson Xavier NX — real-time optimised inference for surveillance, search-and-rescue and emergency response.

Platform
Jetson Xavier NX
Airframe
HolyBro
Runtime
TensorRT
  • Computer vision
  • Embedded systems
  • TensorRT
  • Drone integration
Mission brief →
MSN-02Drone · Mission control

Kiosk ground station software

One-click autonomous drone mission control

A Qt-based ground station with integrated MAVLink communication — operators launch complex autonomous missions in a single click, with minimal technical expertise required.

Framework
Qt
Protocol
MAVLink
Mode
Kiosk UI
  • Qt development
  • UI/UX design
  • MAVLink
  • Autonomous systems
Mission brief →
MSN-03Satellite · Geospatial

AI-powered change detection

Satellite imagery analysis and monitoring

Custom deep learning models that detect change across multi-temporal satellite and aerial imagery — for urban planning, environmental monitoring and geospatial analysis.

Input
Multi-temporal EO
Task
Change detection
Scale
Large datasets
  • Deep learning
  • Model optimisation
  • Computer vision
  • Satellite imagery
Mission brief →

03 / Pipeline

How a mission moves.

  1. CaptureSatellite · UAV · RGB-D · thermal
  2. PerceiveDetection · segmentation · change
  3. ReconstructPoint clouds · SLAM · 3D
  4. OptimisePruning · TensorRT · CUDA
  5. DeployJetson · cloud · ground station

// Open channel

Let's put intelligence in orbit — and in the air.

Launching an AI product, optimising an existing system or exploring research? Let's build something intelligent, scalable and impactful together.