Software Engineer, Backend/Applied ML (Safety & Integrity)

Written by the T&S Network from Character.AI's posting.

Backend and applied ML work on the integrity and safety systems of a consumer AI character platform, covering the human-to-AI interaction risks that come with it. The Safety Engineering team builds detection, prevention and mitigation for traditional integrity issues like spam and content classification, plus newer risks from generative models.

• Spam and platform abuse

What you'd do

• Design, build and maintain scalable backend systems that run integrity and safety features.
• Lead technical design for detecting, preventing and mitigating integrity risks, including harmful or biased generated content.
• Develop and deploy ML models for content classification, including AI-generated content, anomaly detection, risk scoring and behavior analysis.
• Build safeguards for generative AI systems such as content filtering, bias mitigation and output monitoring.
• Work with product, data science, AI research, security and operations teams to define requirements and ship integrity systems.
• Set the technical direction for backend integrity systems and applied ML, weighted toward generative AI safety.
• Mentor engineers and advocate for practices in distributed systems, data engineering and the ML lifecycle.

What it asks for

• Eight or more years of professional software engineering work, mostly on backend systems.
• Expertise in at least one backend language such as Python, Go, Java or C++.
• Demonstrated experience designing and operating large-scale, highly available distributed systems.
• Hands-on applied ML experience on integrity, trust or safety problems.
• Working knowledge of the full ML lifecycle, including validation, A/B testing and operational monitoring.
• Bachelor's, Master's or PhD in computer science, engineering or a related technical field.
• Prior work on a trust and safety, integrity or risk engineering team (preferred).

Our best summary of the posting, and it can be wrong. The employer's posting is what counts: read it in full on Character.AInorth_east before you apply. If we got something wrong, report this listing.