Senior Machine Learning Engineer (AI Agent)

About the Role

We are looking for a Senior NLP Engineer to take full responsibility for ensuring the precision of our information-extraction system. This position involves tackling challenging issues related to extracting material properties and quantitative details from extensive and complex documents like patents and scientific papers, where critical information is dispersed across various sections and the interrelations stretch between paragraphs. This senior role requires hands-on involvement with full ownership over the process: independently designing, prototyping, and implementing algorithm improvements, setting quality measurement standards, and elevating the extraction accuracy to meet production demands.

This role is remote but tied to our Singapore headquarters.

About Us

Patsnap is a rapidly growing, pre-IPO technology company with a mission to revolutionize how intellectual property and R&D teams innovate. Our AI-driven platform enhances the productivity of IP and research teams worldwide, helping over 12,000 clients, including major innovators, accelerate product development. Following a successful $300 million Series E funding round pushing us to unicorn valuation, we maintain a dynamic and committed workforce with a strong entrepreneurial culture spanning global offices.

Your Responsibilities

• Analyze and interpret extensive structured and unstructured data via NLP techniques including data processing, named entity recognition (NER), relationship, and normalization extraction.
• Own and enhance our information-extraction pipeline from model creation to deployment.
• Design scalable document-level extraction solutions capable of handling long documents where entity attributes and their associations cross multiple sections.
• Drive algorithm improvements independently, developing and refining models aligned to business needs.
• Construct hybrid approaches combining NER, rule-based methods, and large language models (LLMs), deploying each method where most effective.
• Collaborate with domain specialists to establish evaluation methods, annotation guidelines, and quality benchmarks.
• Efficiently process and extract information from large-scale document corpora.

Qualifications

• Bachelor’s or Master’s degree in Computer Science or related discipline.
• Minimum of 4 years practical experience in NLP and machine learning engineering with production-level responsibility for model or pipeline quality.
• Demonstrated success with long-document or document-level information extraction and normalization techniques.
• Strong foundational software engineering skills with experience working at scale.
• Expertise in NER and entity extraction methodologies, spanning rule-based, statistical, and LLM-driven techniques.
• Proficient programming skills and familiarity with vibe coding practices.