Job Description

Responsibilities:

• Work with machine learning / deep learning experts to design and implement networks optimised for CPU/GPU-accelerated inference
• Develop advanced inference engine techniques to enable real-time deployment of models
• Deploy productized networks to run efficiently on our hardware (edge based, cloud based, server based)
• Actively involved in MLOps operations of the deployed model.
• Develop tools for profiling, analysing, and improving both training and inference performance

Requirements:

• A minimum of 4 years of relevant experience in a similar role
• Bachelor of Engineering (BTECH / BE) or Masters (MTECH / MS / MCA) in Computer Science / Information Science or Ph. D in Computer Science
• Excellent in machine learning and deep learning concepts
• Data understanding and analysis
• Preprocessing and dimensionality reduction techniques
• Supervised / Unsupervised / semi-supervised learning
• CNN family architectures
• RNN family architectures
• Auto encoders
• GAN and Transformers
• Excellent in Python programming
• Should have good understanding in software design skills
• Intimate understanding and hands on experience of using machine learning and deep learning libraries
• Proven GPU programming (e. g., CUDA, OpenCL) track record
• Should have understanding of unit testing and its test report generation
• Good coding documentation skills
• Good knowledge on problem solving, data structure and Algorithms
• Good communication skills to communicate ideas and issues to the team
• Code contributions to an open-source deep learning community
• Experience using GPU-accelerated libraries (e. g., cuDNN and cuBLAS)
• Experience with code generation & optimisation
• Experience with network model compression and quantisation
• Exposure to TensorRT kinds of tool for DL implementation