Job Description – AI Benchmarking & Performance Engineer

Job Title: AI Benchmarking & Performance Engineer

Experience: 8-12 Years

Location: Bangalore

Education: Bachelor's or Master's Degree in Computer Science, Electronics, Electrical Engineering, AI/ML, Robotics, or related field

Role Overview

We are seeking a highly skilled AI Benchmarking & Performance Engineer to evaluate, analyze, and optimize AI workloads across modern computing platforms including CPU, GPU, and NPU architectures. The ideal candidate will have strong expertise in AI model benchmarking, system-level performance analysis, profiling, and optimization for real-world AI applications spanning computer vision, speech processing, robotics, and edge AI deployments.

This role requires close collaboration with software, hardware, and AI engineering teams to deliver actionable performance insights and drive optimal utilization of computing resources.

Key Responsibilities

Benchmark AI models such as OpenVLA, YOLOv8, Whisper, Pi Droid, Groot, MLPerf Client, and similar AI workloads.
Analyze key performance metrics including:Latency
Throughput (FPS)
Memory Utilization
Power Efficiency
Resource Consumption
Design and execute repeatable benchmark pipelines across multiple hardware platforms.
Perform system-level performance analysis to identify bottlenecks in:CPU
GPU
NPU
Memory Subsystems
Data Movement Pipelines
Optimize AI inference pipelines through profiling and performance tuning techniques.
Integrate AI models into robotics, edge AI, and production application workflows and evaluate end-to-end performance.
Conduct competitive performance comparisons across platforms such as AMD, NVIDIA, Qualcomm, Intel, and ARM-based architectures.
Develop automation frameworks for benchmark execution, result collection, and reporting.
Generate technical reports, comparative analysis, and performance recommendations for engineering and business stakeholders.
Collaborate with cross-functional teams to improve AI deployment efficiency and scalability.

Required Skills

8-12 years of experience in Performance Engineering, System Analysis, or AI Platform Optimization.
Strong understanding of AI/ML model architectures and inference workflows.
Hands-on experience with:PyTorch
ONNX Runtime
TensorRT (Preferred)
OpenVINO (Preferred)
Strong experience in Linux environments and performance analysis tools.
Knowledge of modern:CPU architectures
GPU architectures
NPU/AI Accelerators
Memory hierarchy and system architecture
Experience with profiling and benchmarking tools:perf
Intel VTune
NVIDIA Nsight
ROCm profiling tools
MLPerf benchmarking frameworks
Strong programming skills in:Python
C/C++
Shell/Perl scripting
Excellent debugging, analytical, and problem-solving capabilities.
Experience working with large-scale performance datasets and generating performance insights.

Preferred Skills

Experience with Robotics frameworks such as ROS/ROS2.
Knowledge of Edge AI deployment and optimization.
Experience with simulation environments and validation platforms.
Familiarity with container technologies:Docker
Kubernetes
Experience optimizing AI workloads on embedded platforms.
Understanding of model quantization, pruning, and acceleration techniques.
Exposure to autonomous systems, robotics, or computer vision applications.

Key Technical Skills

AI Benchmarking
Performance Engineering
Machine Learning Inference Optimization
PyTorch
ONNX Runtime
MLPerf
CPU/GPU/NPU Performance Analysis
VTune
Nsight
Linux Performance Tuning
Python
C/C++
ROS / ROS2
Edge AI
System Profiling
Bottleneck Analysis