AI Engineer - Video Analysis Core

Tổng quan về công việc và trách nhiệm

Established for over 20 years in Vietnam with more than 350 members, GIANTY is proud of providing numerous successful services to millions of users worldwide OUR CORE SERVICES: Development Services (Outsource & Offshore), In-House Development (Apps, Games, Anime, Arts, Design, Big Data/Analytics, IoT, VR, AI) Your Mission You will be the founding AI engineer to architect and build the core video analysis engine. The system will combine multi-modal perception (video, audio, text), LLMs, and multi-agent reasoning to understand human behavior, actions, and outcomes from raw video streams. You will work closely with product leaders and domain experts to turn ideas into working features, ensuring our platform adapts across domains while staying fast, accurate, and scalable. - Design and implement the AI pipeline for video understanding, including: - Frame extraction; object and behavior detection - Temporal event segmentation and context tagging - Multimodal fusion (vision, audio, transcripts, sensor data) - Integrate LLMs, RAG, and multi-agent orchestration to interpret, summarize, and reason over video events. - Build real-time or near-real-time inference workflows with a focus on performance and reliability. - Own end-to-end delivery from research prototypes to production-ready APIs and services. - Collaborate with frontend engineers and PMs to define features and deliver “show, not tell” demos weekly. - Optimize models for both edge devices and cloud deployment. - Keep the system domain-agnostic, ensuring adaptability across Education, Retail, Operations, Security, and other verticals.