Applied Computer Vision Engineer

Job Description 

The Video Perception (ViPER) team is responsible for developing cutting-edge solutions for Advanced Driver Assistance Systems (ADAS). Our team is responsible for developing end-to-end solutions by leveraging the latest advancements in the field of computer vision/deep learning. Our goal is to drive the future of autonomous driving systems by providing reliable, high-performance video perception systems.  

We are looking for a talented & driven engineers to join our Retrieval team within ViPER which is responsible for building high-performance systems to ingest and retrieve large-scale driving data, critical for training and validating ADAS models. You will work closely with perception, planning & feature teams to develop intelligent data selection pipelines that power machine learning workflows across various ADAS features.  

The ideal candidate should have experience in building large-scale data systems, efficient querying techniques along with retrieval & recommendation systems, and a strong understanding of deep learning fundamentals. You will deploy tools & services which will be used to retrieve edge-case scenarios, be responsible for developing powerful search and indexing tools, and optimize data pipelines for automating dataset curation for training & validation. 

Requirements 

• Strong technical background: At least a Bachelors/Master of Engineering in Computer Science or Electrical Sciences or related areas with 4 years of industry experience in designing and implementing vision-based algorithms [prior knowledge of working with deep neural networks is a strong plus]. 

• Skilled software engineer with experience in Python/C++ coupled with strong problem-solving skills. 

• Passionate about solving real-world problems: Ideally worked on autonomous driving systems before. 

• A team player: You take ownership and work with the team to deliver exceptional results. 

• Ability to build and iterate quickly (AGILE mindset). You enjoy working in a fast-paced environment, and you are comfortable in all stages of a product development cycle. 

• Good hands on: You like working with production grade machine learning pipelines, from dataset curation all the way till training, validation & deployment. 

• Good communication skills: You are able to explain technical solutions clearly to all stakeholders. You have experience writing clear, concise, and detailed documentation.  

Skills 

• Strong knowledge in DL (3yrs min): Computer vision (object detection, segmentation, depth estimation), Sensor fusion (camera, radar, LiDAR), Supervised & Self-supervised learning & Transfer Learning, CNNs and transformers, Real-time model deployment and optimization, Data augmentation and annotation, Scene understanding and 3D perception, Active learning and Semi-automated data annotation techniques 

• Hands on experience in Full stack Vision/Image related projects in production (3yrs min) 

• Data: Visualization, Data preparation, Data handling, Data analysis & deriving insights 

• Model frameworks: Tensorflow & PyTorch 

• Deployment: Docker, Kubernetes 

• Cloud deployment: Azure/AWS 

• Other skiils: Git, Linux, Cuda, OpenMP, Distributed systems  

B.E/B.Tech

4+ years