Edge AI Specialist - QuantumPulse Technologies

Polluxa, Inc.; Tamil Nadu; On-site; INR 900,000 - 1,300,000; Full time

Job summary

Quantum Pulse Technologies seeks an Edge AI Specialist to translate complex ML models into lightweight AI applications for resource-constrained edge devices. You will work at the SECE Innovation Hub, Coimbatore, collaborating with software, hardware, and robotics teams to deploy real-time computer vision and predictive AI workloads to the edge.

The role emphasizes on-device deployment, hardware-aware optimization, and MLOps for automated edge deployment and monitoring.

Qualifications

• 3–5 years in ML/CV or embedded software with on-device deployment expertise.
• Proficiency in C++ and Python for real time AI pipelines.
• Experience with edge AI frameworks and compilers.

Responsibilities

• Edge model optimization: quantify, prune, compress for low-latency execution.
• On-device deployment to edge accelerators and microcontrollers.
• Benchmark and tune inference pipelines across CPUs/GPUs/NPUs/DSPs.

Skills

C++

Python

On-device deployment

PyTorch

TensorFlow

Tools

TensorRT

TensorFlow Lite

ONNX Runtime

OpenVINO

STM32Cube.AI

Job description

Job Title: Edge AI Specialist

Company Name: Quantum Pulse Technologies

Job Type: Full-Time (On-site / Hybrid)

Experience Level: 3 to 5 Years

Job Overview

Quantum Pulse Technologies is seeking a high-performing Edge AI Specialist to join our team at the SECE Innovation Hub, Coimbatore . In this role, you will be responsible for translating complex deep learning and machine learning models into lightweight, highly optimized AI applications capable of running locally on resource-constrained embedded devices, single-board computers, and custom hardware targets. Located inside the technology hub at Sri Eshwar College of Engineering, you will collaborate with cross-functional software, hardware, and robotics teams to deploy real-time computer vision, signal processing, and predictive AI workloads to the edge.

Key Responsibilities

• Edge Model Optimization: Quantize (INT8/FP16), prune, compress, and compile deep learning models for low-latency, low-power execution on edge hardware.
• On-Device Deployment: Deploy trained models onto edge accelerators and microcontrollers using frameworks like TensorRT, TensorFlow Lite, OpenVINO, and ONNX Runtime.
• Hardware-Software Co-Design: Benchmark, profile, and tune deep learning inference pipelines across CPUs, GPUs, NPUs, and DSPs to meet strict real-time and thermal constraints.
• Data Pipeline &Preprocessing: Build efficient C++/Python data ingestion and preprocessing pipelines for real-time video, audio, and multi-modal sensor streams.
• System Integration: Work closely with embedded system and firmware engineers to integrate Edge AI runtimes into production firmware, Linux environments, or RTOS setups.
• Model Validation & Testing: Monitor accuracy-versus-latency trade-offs on target physical hardware and refine quantization-aware training (QAT) pipelines as necessary.
• Field Deployment &MLOps: Establish automated edge deployment, continuous monitoring, and remote model updating protocols for deployed devices.
• Architected internal project requirements and facilitated high-impact training programs to upskill enthusiastic internal talent.

Required Skills & Qualifications

Core Technical Skills:

• Experience: 3 to 5 years in machine learning, computer vision, or embedded software engineering, with a focus on on-device deployment.
• Programming Languages: Proficiency in C++ (Modern C++14/17) and Python .
• AI/ML Frameworks: Practical experience with PyTorch or TensorFlow for training and fine-tuning neural networks.
• Edge Frameworks & Compilers: Operational experience with TensorRT , TensorFlow Lite , ONNX Runtime , OpenVINO , or STM32Cube.AI .
• Hardware Platforms: Hands-on experience deploying AI on platforms such as NVIDIA Jetson (Nano/Orin) , Google Coral TPU , Raspberry Pi , or NPU/ARM Cortex-based SoCs .