Technical Intern

📋 Description

• • This Technical Intern role at Centific Global Technologies Pte. Ltd. is a Ph.D. Research Intern position focused on advancing perception and embodied intelligence through cutting-edge research in computer vision, multimodal large models, and physical AI. The role matters because it directly contributes to building state-of-the-art Vision AI systems that power scalable, safe GenAI deployment for enterprise clients and the Magnificent Seven, reducing costs and accelerating time-to-market.
• • Day to day, the intern will build and fine-tune models for 2D/3D perception tasks including detection, tracking, segmentation, pose and activity recognition, and scene understanding; train and evaluate vision-language models (VLMs) for grounding, dense captioning, temporal QA, and tool-use; prototype perception-in-the-loop policies that bridge pixels to actions using simulation and real data; curate high-quality datasets and design rigorous evaluation protocols; package research into reproducible services using Kubernetes, Docker, Ray, and FastAPI; and orchestrate multi-agent workflows combining perception, reasoning, simulation, and code generation for self-correction and validation.
• • Centific is a frontier AI data foundry with over 150 PhDs and data scientists, 4,000+ AI practitioners, and 1.8 million vertical domain experts across 230 markets. The company specializes in creating contextual, multilingual, pre-trained datasets, fine-tuned industry-specific LLMs, and RAG pipelines supported by vector databases, enabling zero-distance innovation™ that reduces GenAI costs by up to 80% and accelerates deployment by 50%. The team emphasizes bridging AI creators with industry leaders to unlock business value through scalable, responsible GenAI adoption.
• • The intern will gain hands-on experience translating research into production systems, earn mentorship from Principal Architects and senior researchers, develop end-to-end MLOps skills in modern AI stacks, produce publishable or open-sourced outcomes, and contribute to measurable product KPIs in latency, accuracy, or robustness—building a strong foundation for a career in embodied AI and multimodal systems research.