Software Engineer III, Commerce Actor Safety, Intelligence
Indexed description
Minimum qualifications:
- Bachelor’s degree or equivalent practical experience.
- 2 years of experience with software development in one or more programming languages Python etc, or 1 year of experience with an advanced degree.
- 2 years of experience in machine learning development and Machine Learning (ML) infrastructure (e.g., model deployment, model evaluation, data processing, debugging, or fine tuning) or large-scale data systems.
- Experience with relational databases and writing SQL queries.
- Master's degree or PhD in Computer Science or related technical fields.
- 2 years of experience with data structures and algorithms.
- Experience working with distributed infrastructure, storage, or application frameworks (such as Spanner or server platform).
- Experience or interest in Large Language Models (LLMs), Gemini, prompt engineering, fine-tuning, or graph neural networks/graph mining.
The Commerce Safety team ensures a safe and trustworthy Google Commerce experience. We're on the front lines, protecting users and merchants by maintaining content and merchant integrity and compliance.
As a Software Engineer on the Actor Safety team, you will own and execute end-to-end components across the full-lifecycle development of our abuse intelligence stack including our detection models, agents and moderation platforms.People shop on Google more than a billion times a day - and the Commerce team is responsible for building the experiences that serve these users. The mission for Google Commerce is to be an essential part of the shopping journey for consumers - from inspiration to to a simple and secure checkout experience - and the best place for retailers/merchants to connect with consumers. We support and partner with the commerce ecosystem, from large retailers to small local merchants, to give them the tools, technology and scale to thrive in today’s digital world.
Responsibilities
- Contribute to and execute the actor safety strategy in collaboration with the PM and Eng leads by blending abuse prevention frameworks with targeted, tactical protections. Analyze complex abuse patterns and system precision/recall metrics to drive continuous improvements across our AI models (mScore, potential damage, identity change), detection systems, and operational workflows in close collaboration with partner Trust and Safety teams.
- Design, train, deploy, and iterate on ML models including Gemini LLMs, GNNs, and NLP architectures to detect spam, fraudulent actors, and abusive behaviour across shopping products. Build data pipelines for training data curation, establishing active learning workflows, synthetic datasets, and robust frameworks (LLM-as-a-judge) for model evaluation, monitoring, and performance tracking.
- Design and build high-throughput, resilient back-end services with production hygiene. Build our intelligence platform, managing models and seamlessly integrating with the actor lifecycle.
Create a free Caio profile to unlock more results and save your role and location preferences.
Unlock free search