Researcher, Multimodal Safety

OpenAI
San Francisco
Workplace: HybridFull timeUSD 295,000 - 445,000 annuallyFunction: Research & Scientific (R&D)Skills: ["Safety judgment","Research judgment","Problem diagnosis","Hypothesis generation","Cross-functional collaboration"]

Work on the Chat and Multimodal Safety team to advance multimodal safety research for text, vision, and audio. Define safety research directions, and build training and evaluation methods for VLMs, including post-training, safety evals, and interventions that improve safe responses across contexts. Collaborate with Personal AGI, Consumer Devices, and product/model teams to translate research into safer ambient and personalized multimodal experiences.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
OpenAI
OpenAI
2 days ago

Researcher, Multimodal Safety

āœ“ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 6 hours agoStatus: Live

Job Summary

Work on the Chat and Multimodal Safety team to advance multimodal safety research for text, vision, and audio. Define safety research directions, and build training and evaluation methods for VLMs, including post-training, safety evals, and interventions that improve safe responses across contexts. Collaborate with Personal AGI, Consumer Devices, and product/model teams to translate research into safer ambient and personalized multimodal experiences.
Location: San Francisco
Workplace: Hybrid
Employment Type: Full time
Job Function: Research & Scientific (R&D)
Seniority: Mid level

Key Responsibilities

  • ā€¢Define and advance multimodal safety research for text, vision, and audio, connecting perception and semantic understanding to safe model behavior.
  • ā€¢Build training and evaluation methods for VLMs, including post-training, safety evals, and interventions that support safe responses in varied contexts.
  • ā€¢Collaborate with Personal AGI, Consumer Devices, and product/model teams to translate research into safer ambient, embedded, and personalized multimodal experiences.

Pay and Benefits

Salary: USD 295,000 - 445,000 annually
Equity and Bonus:Equity
Perks:Relocation

Key Requirements

  • ā€¢Track record building or advancing multimodal models, with depth in vision-language models, video understanding, image generation, audio, or multimodal reasoning.
  • ā€¢Understand multimodal systems end to end, including encoders, projection layers, modality fusion, cross-modal reasoning, scaling, and inference tradeoffs.
  • ā€¢Experience improving frontier model behavior through post-training, including methods such as SFT, RL, data curation, synthetic data, evaluation, and error analysis.
  • ā€¢Strong research and engineering judgment for open-ended safety problems, including forming testable hypotheses and diagnosing model failures.
  • ā€¢Ability to translate research findings into robust improvements to model safety.
Experience:MultimodalVision-language modelsPost-trainingAI safety
Skills:Safety judgmentResearch judgmentProblem diagnosisHypothesis generationCross-functional collaboration
Tech Stack:VLMsVision-language modelsModality fusionImage encodersPost-trainingSafety evalsInterventionsSFTRLSynthetic dataError analysisEncodersProjection layersCross-modal reasoning

Company Brief

OpenAI
Develops and deploys advanced generative AI models (including ChatGPT and DALLĀ·E) and AI infrastructure, providing APIs and consumer products to accelerate safe AGI for broad benefit.
Industry: AI & Machine Learning
Company Size: Enterprise (1,001+ employees)
Revenue: USD 1B+
Growth: Scaleup
Valuation: Hectocorn (USD 100B+)
Funding: Series E+
Headquarters: San Francisco, United States
Founded: 2015
Glassdoor
Glassdoor: 4.4
WebsiteLinkedInGlassdoor