Gen AI Multimodal SME

Summary

Join SweetRush's Custom Learning Project team as a Subject Matter Expert (SME) to help create a new certification course on Google's Gemini and DeepMind technologies. This fully remote position offers flexibility, working within US business hours. You will collaborate with learning experience designers and content developers to translate real-world applications into accessible learning materials for technical professionals. The role requires deep experience in AI/ML, Google Cloud, and familiarity with multimodal systems and the DeepMind ecosystem. The hourly rate is negotiable. SweetRush is a culture-forward team that values diversity, respect, and caring.

Requirements

• Demonstrated hands-on experience building, deploying, and maintaining sophisticated AI-powered applications using Gemini APIs/SDKs within the Google Cloud ecosystem, especially in Firebase Studio and VS Code
• Proficiency in designing and implementing agent-like application patterns, including multi-turn conversational flows, state management, and complex prompting strategies (e.g., Chain-of-Thought, few-shot, zero-shot)
• Experience integrating Gemini with Google Cloud services (Firestore, Cloud Functions, App Hosting) and external APIs for robust, production-ready solutions
• Proven ability to engineer applications that process, integrate, and generate content across multiple modalities (text, images, audio, video, code) using Gemini’s native multimodal capabilities
• Skilled in building and orchestrating pipelines for multimodal data handling, synchronization, and complex interaction patterns within application logic
• Experience designing and implementing production-grade RAG systems, including integration with vector databases (e.g., Pinecone, ChromaDB) and engineering data pipelines for indexing and retrieval
• Ability to manage agent state, memory, and persistence for multi-turn and long-running interactions
• Proficiency leveraging AI-assisted coding features in Firebase Studio (chat, inline code, command execution) and using App Prototyping agents or frameworks like Genkit for rapid prototyping and structuring agentic logic
• Strong command of modern development workflows, including Git/GitHub, code reviews, and collaborative development practices
• Experience designing scalable, fault-tolerant deployment architectures for multimodal and agentic AI applications using Firebase App Hosting, Cloud Run, or similar serverless/cloud platforms
• Advanced MLOps skills, including monitoring, logging, alerting, and versioning for generative AI systems and agents
• Deep understanding of security best practices: prompt injection mitigation (across modalities), secure API key management, authentication/authorization, and data privacy
• Demonstrated ability to engineer for responsible AI, including bias detection, fairness, transparency, and implementation of safety mechanisms in agentic and multimodal applications
• Experience addressing ethical challenges in the deployment and operation of advanced AI systems
• Proven success designing, reviewing, and delivering advanced, project-based curriculum and hands-on labs for experienced software developers and engineers
• Ability to translate complex engineering concepts (RAG, multimodal integration, agentic patterns, MLOps, security, responsible AI) into clear, actionable learning materials and real-world projects
• 5+ years of professional experience in AI-powered application development, with a focus on generative and multimodal AI
• Strong programming skills in Python and JavaScript/TypeScript; experience with modern frameworks and cloud-native development
• Bachelor’s or Master’s degree in Computer Science, Data Engineering, AI, or a related technical field
• Ability to explain advanced technical concepts (e.g., fusion transformers, multimodal embeddings, RAG workflows) to learners in an accessible way
• Strong programming experience in Python and experience deploying machine learning pipelines
• Ability to work independently, take ownership of deliverables, and collaborate closely with designers and project managers

Responsibilities

• Translate real-world Gemini and DeepMind applications into accessible, hands-on learning for technical professionals
• Guide the creation of labs and projects that allow learners to build pipelines for image-text fusion, deploy Gemini APIs, and experiment with DeepMind’s reinforcement learning libraries
• Contribute technical depth across activities, from high-level course structure down to example code, diagrams, voiceover scripts, and data pipelines
• Ensure all content reflects current, accurate usage of Google’s multimodal tools and services
• Be available during U.S. business hours to support project milestones, reviews, and content feedback

Preferred Qualifications

• Experience with Google DeepMind tools (JAX, Haiku, RLax, DeepMind Control Suite/Lab) and reinforcement learning pipelines
• Familiarity with open data formats (Delta, Parquet, Iceberg) and scalable data engineering practices
• Prior contributions to open-source AI projects or technical community engagement

Benefits

Remote work, flexible hours