Software Engineer, Multimedia & Multimodal AI

Meta’s Applied AI organization develops and evaluates AI models, and its MultiMedia & MultiModality team works across image, video, audio, speech, and music. The Software Engineer will define multimedia data and evaluation approaches, build reliable evaluation systems and scalable production pipelines, and work closely with researchers to improve model capabilities.

Responsibilities

• Design, create, and quality-review expert multimedia tasks and reference outputs for model training and evaluation
• Define task guidelines, rubrics, and quality criteria
• Calibrate reviewers to apply them consistently
• Build and harden evaluations
• Make graders reliable, separate model failure from instrumentation failure, and reproduce and debug quality issues to resolution
• Analyze failure modes in model outputs and propose new task types or data to close gaps
• Design and build agentic workflows and pipelines, including human-in-the-loop and expert-in-the-loop designs, to automate data production and scale output past what manual authoring supports
• Mentor engineers on the team, contribute to hiring and onboarding, and raise the bar on evaluation and quality practice

Preferred

• Hands-on experience evaluating or red-teaming multimodal models, or creating the data used to improve them
• Experience designing benchmarks or evaluations for model capability, with attention to grading reliability, reproducibility and label quality
• Working knowledge of common video, audio and streaming standards (e.g. H.265, MPEG-DASH, WebRTC)
• Fluency with professional media tooling (e.g. Premiere, After Effects, Blender, Pro Tools). We care about this because authoring tasks a model cannot solve requires knowing what expert media work actually looks like
• Experience building data pipelines for image, video, audio, speech or complex media formats, including versioning, lineage and provenance
• Experience designing AI agents, orchestration, or human-in-the-loop systems
• Experience working directly with researchers and translating research needs into engineering and evaluation work
• Understanding of Responsible AI practices and building quality controls into AI output
• Experience with zero-to-one work: forming a charter and standing up process while priorities are still moving
• Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements)
• Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews)
• Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies
• Multimodal Model Evaluation
• Model Red-Teaming
• Training Data Creation
• Benchmark Design
• Model Capability Evaluation
• Grading Reliability
• Reproducible Evaluations
• Label Quality Assessment
• Video Standards
• Audio Standards
• Streaming Standards
• H.265
• MPEG-DASH
• WebRTC
• Adobe Premiere
• Adobe After Effects
• Blender
• Pro Tools
• Professional Media Production
• Media Task Authoring
• Multimedia Data Pipelines
• Data Versioning
• Data Lineage
• Data Provenance
• AI Agent Design
• Agent Orchestration
• Human-in-the-Loop Systems
• Research-to-Engineering Translation
• Evaluation Engineering
• Responsible AI
• AI Quality Controls
• Process Design
• Workflow Optimization
• AI Tool Integration
• Impact Measurement
• Prompt Engineering
• Context Engineering
• Risk Assessment
• Bias Mitigation
• AI Output Reviews
• Quality Reviews
• Accuracy Reviews
• Emerging AI Technologies
• Researcher Collaboration
• Adaptability
• Initiative
• Continuous Learning

Benefits

• Bonus
• Equity

Company

• Social networking and immersive augmented/virtual reality technologies.
• Menlo Park, California, USA
• https://www.meta.com/
• Founded in 2004
• 10001+ employees

Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
• 8+ years of software engineering experience, or a PhD plus 5 years, including significant depth in one or more media modalities (video, image, audio, speech, or music)
• Demonstrated experience designing and building multimedia pipelines, workflows or tooling
• Working understanding of how models are trained and evaluated, and of how data quality and coverage shape model behavior
• Experience owning software components or systems end to end, and driving work with cross-functional partners
• Hands-on experience evaluating or red-teaming multimodal models, or creating the data used to improve them
• Experience designing benchmarks or evaluations for model capability, with attention to grading reliability, reproducibility and label quality
• Working knowledge of common video, audio and streaming standards (e.g. H.265, MPEG-DASH, WebRTC)
• Fluency with professional media tooling (e.g. Premiere, After Effects, Blender, Pro Tools). We care about this because authoring tasks a model cannot solve requires knowing what expert media work actually looks like
• Experience building data pipelines for image, video, audio, speech or complex media formats, including versioning, lineage and provenance
• Experience designing AI agents, orchestration, or human-in-the-loop systems
• Experience working directly with researchers and translating research needs into engineering and evaluation work
• Understanding of Responsible AI practices and building quality controls into AI output
• Experience with zero-to-one work: forming a charter and standing up process while priorities are still moving
• Demonstrated ability to integrate AI tools to optimize/redesign workflows and drive measurable impact (e.g., efficiency gains, quality improvements)
• Experience adhering to and implementing responsible, ethical AI practices (e.g., risk assessment, bias mitigation, quality and accuracy reviews)
• Demonstrated ongoing AI skill development (e.g., prompt/context engineering, agent orchestration) and staying current with emerging AI technologies

Required skills

• AI Agent Orchestration, Adobe After Effects, Adobe Premiere, Blender, H.265, Human-in-the-Loop Systems, MPEG-DASH, Model Training and Evaluation, Multimedia Pipeline Engineering, Multimodal Benchmark Design, Multimodal Model Evaluation, Multimodal Model Red-Teaming, Pro Tools, Responsible AI, WebRTC

Experience

• 8+ years