Algorithm Test Development Engineer

Company Introduction PatSnap is a a SaaS-based tech innovation intelligence service provider. Patsnap helps R&D leaders to maximise the value of innovation intelligence within their R&D workflow and strategic planning. As the global leaders in connected innovation intelligence, PatSnap use AI-powered and machine learning technology to comb through billions of datasets, and help innovators connect the dots. Patsnap recently completed a Series E funding round of $300 million, led by Tencent and the SoftBank Vision Fund II. This represents the largest round of funding in the SaaS sector since 2020, Patsnap is a unicorn in the SaaS track.

Department Introduction The R&D SG Department is dedicated to building leading capabilities in natural language processing technology and services. This is achieved by leveraging technologies such as Deep Learning, NLP, CV, and Knowledge Graph to create top-tier products.

Responsibilities

• Responsible for the design and implementation of AI-related business needs (such as Material PUM NER etc.) evaluation;
• Responsible for the corpus construction and test capability assessment of AI-related engine capabilities;
• Responsible for the construction of the algorithm test environment and platform;
• Algorithm evaluation, including requirement analysis, test plan, corpus construction, design indicators based on production needs and coding, defect tracking and quality analysis, etc.;
• Responsible for establishing a system for subjective and objective evaluation of algorithm quality;
• Actively cooperate with the data team and r&d team to optimize the process and methods of testing, data collection, and annotation, and to improve the quality and efficiency of the team's work;

Requirements

• Proficient in at least one of Python, C++, Java, and other programming languages, proficient in database languages such as SQL, etc., familiar with the Linux operating system, and have a solid programming foundation;
• Familiar with mainstream automated testing open-source frameworks and tools, and can maintain and optimize these frameworks or tools;
• Have experience in AI product/algorithm testing, proficient in algorithm quality evaluation;
• Familiar with the principles of deep learning and common machine learning algorithms, and can use clustering, classification, regression, ranking and other models to solve problems;
• In-depth understanding and application of common language model evaluation indicators such as Cross-entropy, BPC/BPW, Perplexity, ROUGE, etc.;
• Strong insight and practical ability to learn new technologies;
• Familiar with the software development lifecycle.