Vision and Imaging - Intern Intern

Tonbo Imaging; Bengaluru; On-site; INR 1,200,000 - 1,800,000; Full time

Job summary

Tonbo Imaging in Bengaluru seeks a ML/Software engineer to advance deep learning models for object detection, classification and tracking on EO/IR imagery. You will optimize models for edge deployment on NVIDIA Jetson platforms using TensorRT, ONNX, and TFLite, and drive real-time inference pipelines and performance tuning.

You will curate EO/IR datasets, run structured experiments to track model quality, and collaborate with systems, embedded, and product teams to move prototypes toward fielded

Qualifications

• Experience developing DL models for EO/IR imagery.
• Experience with real-time inference on edge devices is desirable.
• Proficiency in evaluating and evolving computer vision models.

Responsibilities

• Model development: train, fine-tune, and evaluate deep learning models for object detection, classification, and tracking on EO/IR imagery.
• Edge deployment: optimize and deploy models to NVIDIA Jetson and other edge AI accelerators using TensorRT, ONNX, and TFLite.
• Real-time pipelines: build and maintain multi-camera streaming and inference modules.
• Datasets & evaluation: curate EO/IR datasets, build evaluation harnesses, and run experiments to track model quality over time.
• Cross-functional work: collaborate with systems, embedded, and product teams to take a model from research prototype to a fielded product.
• Experience with hands-on development of code for real-time scenarios beyond PC/Cloud.

Skills

Deep learning

Object detection

Tracking

Model deployment

Edge AI

Cameras

Cross-functional collaboration

Tools

TensorRT

ONNX

TFLite

NVIDIA Jetson

Python

Job description

• Model development: Train, fine-tune, and evaluate deep learning models for object

detection, classification, and tracking on EO and IR imagery.

• Edge deployment: Optimize and deploy models to NVIDIA Jetson and other edge AI

accelerators using TensorRT, ONNX, and TFLite. Profile, quantize, and squeeze real-time performance out of constrained hardware.

• Real-time pipelines: Help build and maintain multi-camera streaming and inference

modules.

• Datasets & evaluation: Curate and augment EO/IR datasets, build evaluation

harnesses, and run structured experiments to track model quality over time.

• Prior experience handling and using Cameras would be great.
• Cross-functional work: Collaborate with systems, embedded, and product teams to take a model from research prototype to a fielded product.

Other Competencies:

• Experience with hands-on development of code for real time scenarios, not limited to PC/Cloud related developments.