The fitness industry is shifting from “experience-driven” to “data-driven” paradigms. In this context, Hang Chen approaches the field from an operator’s perspective, deploying AI and big data across gym operations and instructional services to amplify user value and business performance through scalable, reproducible methods. He began with professional content creation on YouTube and Bilibili, accumulating more than 100,000 subscribers in two years and gaining a clear view of the limits of online science communication: knowledge can be disseminated, but individual differences are difficult to address with sufficient granularity.
At the same time, the Chinese community in the greater Seattle area (Bellevue) long lacked an offline space that combined professionalism, structure, and a sense of belonging for strength training and bodybuilding. Addressing these two pain points, he founded Tiger Gate Gym in August 2023, positioning it as “a bodybuilding community underpinned by an intelligent training system.” The goals are straightforward: make training safer, more effective, and reproducible; and make gym operations more measurable, scalable, and replicable.
Hang Chen is not a programmer pursuing R&D for its own sake; he is a hands-on operator oriented toward business outcomes. The two studies he led both target the core objectives of “improving training quality and enhancing operational efficiency.”
The first, “A Human Pose Recognition Framework Based on YOLOv7-Pose: Applied to Competitive Aerobics,” centers on the question of whether movements are performed correctly. On the gym side, it has been encapsulated as a “movement-quality monitoring and correction” module: for compound lifts such as squats, deadlifts, and bench presses, it recognizes joint angles and movement trajectories in real time, helping coaches quickly identify risky patterns and enabling learners to establish proper techniques more rapidly. The business-level effects are evident—novices ramp up faster, injury risk declines, each coach can serve more clients, and word-of-mouth and membership renewals become more robust.
The second, “Intelligent Fitness Data Analysis and Training Effect Prediction Based on Machine Learning Algorithms,” addresses whether training is effective and how to make it more so. Integrating data from wearable devices, it identifies key factors with greater impact on training outcomes (e.g., body-fat percentage, maximum heart rate, age) and formulates a closed-loop of “assessment—prescription—feedback—review,” including an optimal range for single-session duration (approximately 1.2–1.6 hours). On the operations side, it directly supports personalized program design, group-class scheduling, and recovery strategies, which in turn has significantly improved course satisfaction and member retention.
After these capabilities were integrated into the gym, the operational returns became clear: Tiger Gate Gym now has more than 500 members, with over 1,500 cumulative trial visits. Revenue and repeat purchases are on a sustained upward trajectory, and the business model is stabilizing, laying a data and process foundation for future site selection and replication.
To test the transferability of these methods across populations and scenarios, Hang Chen also conducted related work titled “An Intelligent Physical Training System in Football Education,” transplanting the “indicator selection—layered modeling—effect evaluation” approach to team sports and outdoor conditioning. The purpose is not “cross-domain showmanship,” but to prepare for expanded offerings in corporate team-building, small-group classes, youth athletic camps, and specialized Strength & Conditioning (S&C). Next, the gym will incorporate outdoor and team conditioning into regular operations so that “indoor equipment–based training + outdoor conditioning” form complementary product lines.
Currently, Tiger Gate Gym plans to open two to three additional locations in King County and surrounding areas by 2027, maintaining service density through a model that combines data-driven site selection with a mix of asset-light and asset-heavy deployment. In parallel, the gym is pursuing cooperation with major bodybuilding competitions in Washington State (e.g., the NPC Emerald Cup) to integrate training and contest-preparation resources. On the corporate-health side, it is exploring preferential and customized solutions for employees at technology companies via platforms such as Wellhub, with the aim of shaping standardized B2B (business-to-business) service packages. The plan also includes launching a proprietary supplement line linked to intelligent training prescriptions and iterating lightweight digital tools (member assessment, training logs, and progress dashboards) to support scaled operations.
From validating online traffic, to validating the brick-and-mortar model, and now validating a replicable operating framework, Hang Chen has pursued an intelligence-driven path anchored in business performance. Tiger Gate Gym has become one of the core communities for Chinese bodybuilding and strength training in the Seattle area, while also serving as a commercial case of AI that is “deployable and replicable” in fitness settings. As he often notes: “Goals should be adjusted in a timely manner, and one’s mindset should be continuously iterated. As you accumulate strength, your moment will come.” Going forward, the integrated strategy of “AI + fitness + scenario expansion” will continue to deliver scientific rigor and operational efficiency to more trainees and more cities.
