Senior Research Engineer

Lightspeed Studios; Singapore; On-site; SGD 70,000 - 100,000; Full time

Job summary

Lightspeed Studios in Singapore is looking for a researcher to optimize content generation models, focusing on various modalities. You'll support the productization of algorithms and enhance user experiences while conducting algorithm training.

The ideal candidate has a PhD in Computer Science, strong foundations in machine learning, and fluency in English and Mandarin. Join us to work on cutting-edge technology in a diverse and inclusive environment.

Qualifications

• Strong foundation in algorithm publications preferred.
• Hands-on experience in NLP or multimodal learning is a strong plus.
• Fluency in English and Mandarin required.

Responsibilities

• Research and optimize content generation models.
• Support productization of algorithms.
• Conduct algorithm training and optimization.
• Deploy multimodal generative models in industry.

Skills

Computer vision

Machine learning algorithms

Deep learning fundamentals

Proficient with AIGC frameworks

Teamwork

Excellent communication skills

Education

PhD in Computer Science or related fields

Job description

Job Overview

Business Unit Technology Engineering Group (TEG) supports the company and its business groups on technology and operational platforms, including the construction and operation of R&D management and data centers. TEG operates the largest networking, device, and data center in Asia and leads Tencent’s Technology Committee in strengthening infrastructure R&D through open source collaboration, new platform construction, and business innovation.

Responsibilities

• Research and optimize content generation models (text, image, audio, 3D models, etc.) to improve generation quality, diversity, controllability, and efficiency.
• Support the productization of algorithms and improve user experience and production effectiveness.
• Conduct algorithm training and optimization in areas such as image generation, multimodal large models, and few‑shot learning.
• Accelerate model performance and resolve bottlenecks for real‑world business scenarios through prompt optimization, model R&D, adapter development, and performance acceleration.
• Deploy multimodal generative models in industry and explore model design and optimization in an R&D context.

Qualifications

• PhD (preferably full‑time) in Computer Science, Artificial Intelligence, Mathematics, or related fields.
• Strong foundation in computer vision or machine learning algorithms; publications in top conferences or journals preferred.
• Proficient in machine learning and deep learning fundamentals and familiar with mainstream AIGC frameworks (GAN, VAE, VQGAN, diffusion models, etc.).
• Knowledge of generation model extensions such as ControlNet, LoRA, and Text Inversion.
• Experience with multimodal models like CLIP, ERNIE‑ViL, and other transformer‑based cross‑modal representation models.
• Hands‑on experience in NLP, multimodal learning, or AI‑generated content is a strong plus.
• Strong learning ability, clear logical thinking, excellent communication skills, high curiosity, teamwork, and interpersonal communication skills.
• Fluency in English and Mandarin to work with international and HQ stakeholders.

Equal Employment Opportunity

As an equal‑opportunity employer, Tencent values diversity and strives to support all employees in achieving individual and collective goals.