Research Engineer Intern, Speech Generation - Spring 2026

Research Engineer Intern, Speech Generation - Spring 2026 Genies is an avatar technology company powering the next era of interactive digital identity through Smart Avatars. With the Avatar Framework and intuitive creation tools, Genies enables developers, talent, and creators to generate and deploy game-ready Smart AI companions. The company’s technology stack supports full customization, AI-generated fashion and props, and seamless integration of user-generated content (UGC). Backed by investors including Bob Iger, Silver Lake, BOND, and NEA, Genies’ mission is to become the visual and interactive layer for the LLM-powered internet. Genies is an avatar technology company powering the next era of interactive digital identity through AI companions. With the Avatar Framework and intuitive creation tools, Genies enables developers, talent, and creators to generate and deploy game-ready AI companions. The company’s technology stack supports full customization, AI-generated fashion and props, and seamless integration of user-generated content (UGC). Backed by investors including Bob Iger, Silver Lake, BOND, and NEA, Genies’ mission is to become the visual and interactive layer for the LLM-powered internet. About the Role: Genies is looking for a passionate Research Engineer Intern to join our core AI team at our Bay Area office in San Francisco, CA. You will solve fundamental technical challenges in creating socially-aware, autonomous, and creative AI characters and companions. Working closely with a dedicated team of engineers and researchers, you will bridge the gap between cutting-edge research and real-world product impact. What You'll Be Doing: - Research, design, and implement state-of-the-art Generative AI models for speech generation, including Text-to-Speech (TTS) and Voice Clone. - Investigate and apply novel architectures such as Flow Matching, or Autoregressive Transformers to improve audio fidelity and voice cloning capabilities. - Develop techniques for disentangled representation learning to separate speaker identity from prosody and emotion, enabling highly controllable and expressive character voices. - Optimize deep learning models for real-time inference, focusing on latency reduction and throughput improvement. What You Should Have: - Currently pursuing a Master’s or PhD degree in Computer Science, Machine Learning, Electrical Engineering, or a related quantitative field. - Demonstrated experience with deep learning frameworks (e.g., PyTorch) and Python programming. - Solid understanding of digital signal processing (DSP) fundamentals and audio synthesis concepts. - Experience with at least one relevant area: Speech Synthesis, Voice Clone, or Generative Modeling. What We Prefer - First-author publications in top-tier venues such as ICASSP, Interspeech, NeurIPS, ICML, or ICLR. - Familiarity with model optimization techniques (quantization, distillation) and inference acceleration. - Knowledge or experience integrating machine learning models into production environments. - Previous internship or research experience in audio AI or related domains. How Genies will support you Genies is a well-funded, growing start-up that cares deeply about each of our employees' growth and success. Our roles and their responsibilities are created with a breadth of scope that introduces each employee to exciting new challenges and opportunities that a growing start-up encounters. The actual base pay is dependent upon a number of factors, including: professional background, training, transferable skills, work experience, education, location, business and product needs, and market demand. The base pay range is subject to change and may be modified in the future. Full-time employees may also be eligible for equity compensation, in addition to a robust health, wellness, and benefits package. Starting Salary Range: $45-$55 Here's why you'll love working at Genies: - You'll work with a team that you’ll be able to ... (truncated, view full listing at source)