SK Broadband is innovating its network quality management by building an environment where any employee can directly create AI agents and put them to work.
In February this year, SK Broadband built the "Playground" platform, equipped with network data analysis and coding support functions, through collaboration between its internal network organization and its AT·DT Center. The platform automated the setup of development environments that previously took more than two months, sharply lowering the entry barrier so that development can begin in just five minutes.
Previously, employees faced high technical barriers to introducing data analysis or automation into their own work. They could only use AI in the field after personally handling complex procedures such as applying for server access permissions and setting up development environments.
Playground demonstrates synergy by linking with "LDAS," a location-based internal data analysis system. Employees can immediately use a variety of internal data, including network equipment, quality, and traffic data, as well as customer experience indicators, for AI agent development.
The number of AI applications being developed and operated through Playground reaches about 600. Among them, some 30 AI agents that judge situations and carry out actions on their own have been applied in the field. A representative example is the AI monitoring and diagnostic agent "C-One," in which AI automatically detects abnormal signs in wired networks based on customer experience indicators and immediately identifies causes and inspection priorities. The AI comprehensively analyzes abnormal signs such as changes in quality scores, complaint status, and optical signal strength to quickly narrow down the problem section, and even suggests actions to be executed immediately, such as remote equipment resets or on-site optical line inspections. It can also find points requiring inspection, automatically generate reports, and send them to the person in charge.
SK Broadband plans to upgrade "C-One" into an autonomous recovery agent in which AI performs everything from fault detection to handling and recovery on its own. In addition, the company is operating a system in which servers deployed nationwide send signals directly to subscriber devices to measure Wi-Fi and high-speed device quality in real time. The aim is to provide stable communication services by preemptively resolving fault factors, based on AI-driven detection of abnormal signs and rapid recommendation of measures before customer inconvenience occurs.
"As employees create AI agents themselves and apply them in the field, the way we work is fundamentally changing," said Sung Jin-soo, head of SK Broadband's Network Center. "We will further spread the AI development culture and continuously improve network quality."
