Manufacturers spend up to 40% of their sales revenue due to cost of poor quality (COPQ). This includes cost of preventing and detecting product failures. The later a defect is detected, the more resources has been wasted on a defective part. In the worst case, defects could escape their factories. This could result in product recalls that could bring them out of business.
A single log point (out of hundreds) could produce 50GB of data every day. There is a great value to this data as it can tell manufacturers why failures happen. But no human is capable of reviewing this mass of data.
We are building an AI that understands manufacturer's data to reduce the cost of poor quality. The dream is to fix defect causes before they even happen.
Our target customers are high volume manufacturing firms, starting with solar panel manufacturers. The industry is adoptive and has deep pockets. Moreover, we are confident to dominate this exponentially growing 5B$ niche market.
We are looking for our first lead Machine Learning Engineer to turn the dream of a self-aware factory into reality. You will be working hand-in-hand with our CTO (http://github.com/mindworker) and your mission is to build and grow the data platform that manufacturers of the future are going to use.
Your responsibilities can range from:
- Build surface defect detection models that are able to analyze 600 million parts every year
- Build an Outlier Detection model and help hundreds of process engineers to react to issues in the manufacturing line quickly
- R&D for a Defect Forecast Model and dramatically reduce industrial waste
- Design the Data Processing Pipelines that can transform, merge and clean terabytes of data
- Lead and grow your own machine learning team
- 3+ years of experience in software development with machine learning and/or computer vision
- Experience in computer vision - specific skills in object detection, and classification is a plus
- Experience in learning algorithms (decision trees, support vector machines and deep learning) - specific skills in any MaskRCNN, Autoencoders, or Restricted Boltzmann Machine is a plus
- Good communication skills in English (Mandarin is a plus)