Beijing Kaikong Selected for ABB Startup Challenge 2026

Beijing Kaikong Technology Co., Ltd. has been selected as a participant in the ABB Startup Challenge China 2026, the annual innovation competition hosted by global industrial automation leader ABB.

A New Kind of Industrial Development Platform

In July 2026, ABB launched its Startup Challenge China 2026, inviting startups across industrial automation, robotics, AI, and energy to present breakthrough solutions. After a rigorous evaluation process, Beijing Kaikong Technology emerged as one of the selected participants — recognized for its AI-native industrial automation development platform.

This is not a conventional PLC programming IDE with an AI chatbot bolted on. Kaikong’s flagship product, kVPAC IDE, is architected from the ground up to embed AI capabilities into industrial PLC programming workflows.

What Makes kVPAC IDE Different

The industrial automation software market has long been dominated by platforms like CODESYS and Siemens TIA Portal. While these tools are mature and reliable, they were designed in an era before large language models. Kaikong’s approach is fundamentally different:

42+ AI tools, productized for engineering workflows. From natural language code generation in IEC 61131-3 (Structured Text, Ladder Diagram, Function Block Diagram) to intelligent error diagnosis and automated test case generation, each AI capability is engineered for direct integration into daily development tasks. An engineer types a natural language description; the system generates IEC-compliant code. This is not a prototype — these tools are production-ready and validated in real engineering projects.

MCP Server dual-end architecture. Built on the open Model Context Protocol (MCP), Kaikong implements MCP Servers at both the IDE and Runtime (RTE) layers. This enables AI tools to not only read source code but also perceive real-time PLC runtime state — creating a closed-loop “sense-reason-act” cycle that was previously impossible. The dual-end design ensures that AI assistance is not limited to the editing phase but extends into deployment and runtime monitoring.

Rust + WASM high-performance engine. Replacing the sluggish compiled-based IDEs of the past, Kaikong’s Rust-powered engine combined with WebAssembly delivers compilation speeds several times faster while maintaining industrial-grade stability. Engineers no longer need to choose between speed and reliability.

Why This Matters Now

Two structural shifts are converging in industrial automation:

Accelerated localization. As governments worldwide push for domestic industrial software capabilities, market demand for locally built PLC programming platforms is surging. Legacy tools like CODESYS and TIA Portal have long held dominant positions, but the window for domestic alternatives is opening rapidly. Kaikong’s kVPAC IDE is positioned to meet this demand with a production-grade, AI-native solution — not a stopgap, but a platform designed to compete head-to-head with established players.

AI industrialization. Large language models are migrating from consumer internet to industrial domains. But industrial requirements — real-time performance, reliability, safety constraints — demand purpose-built AI solutions rather than repurposed consumer tools. A chatbot that can write Python scripts cannot directly generate IEC 61131-3 Structured Text with industrial-grade correctness guarantees. Kaikong’s architecture is designed specifically for these constraints, with AI tools that understand industrial protocols, safety standards, and real-time execution requirements.

The intersection of these two trends is Kaikong’s core battleground — and exactly where ABB’s innovation agenda is focused. We believe the next three to five years will see a fundamental paradigm shift in industrial automation development tools: from “engineers adapting to tools” to “tools understanding engineers.”

AI-Native Is Not Hype — It’s Verifiable Efficiency

In an era where AI buzzwords saturate every press release, we prefer to speak in data. In real-world engineering tests, kVPAC IDE has demonstrated:

• Code generation: From natural language requirement descriptions to IEC 61131-3 code, average time reduced from hours to minutes
• Error diagnosis: AI-driven intelligent diagnostics cut common compilation error identification time by approximately 60%
• Test automation: Automated test case generation covering common boundary conditions, reducing manual test writing effort significantly

More importantly, these AI tools are not isolated “magic buttons” but deeply embedded organic components of the engineering workflow. When an engineer writes code, AI provides autocomplete; when compilation fails, AI automatically analyzes the error; when runtime anomalies occur, AI combines real-time data to offer diagnostic suggestions. Every step remains under the engineer’s control — explainable, traceable, auditable. This is the non-negotiable baseline requirement for AI in industrial settings.

Strategic Alignment with ABB

ABB’s business spans robotics, motion control, process automation, and electrification. Kaikong’s technical capabilities complement ABB’s ecosystem across multiple dimensions:

• Open protocol stack (DAP/LSP/MCP) enables seamless integration with existing ABB systems, lowering the barrier for ecosystem onboarding
• Full IEC 61131-3 language support (ST/LD/FBD/SFC/IL) ensures compatibility with ABB PLC hardware, so engineers don’t need to switch toolchains
• Cross-platform deployment (Windows/macOS/Linux) provides flexibility for ABB’s global customer base, accommodating different regional IT infrastructure requirements

This competition entry is not just a technology showcase — it marks the start of a deeper engagement with ABB’s startup and innovation ecosystem. Through the ABB Startup Challenge program, we aim to bring Kaikong’s innovation capabilities together with ABB’s global channels, industry influence, and ecosystem resources to help advance the next generation of industrial automation development tools.

What’s Next

The Grand Final of ABB Startup Challenge China 2026 will take place October 27-28, 2026 in Nanjing, where Kaikong will present live demonstrations of:

• 42+ AI tools in action: Natural language PLC code generation, intelligent debugging, automated test case generation
• MCP Server dual-end architecture: Coordinated workflow between IDE and RTE layers
• Full IEC 61131-3 language support: From Structured Text to Ladder Diagram, covering all engineering scenarios
• Real-world engineering cases: Deployment experience and technical results from actual projects

We look forward to engaging with industry peers in Nanjing and exploring the future of deep AI-industrial automation integration together.

About Beijing Kaikong Technology

Beijing Kaikong Technology Co., Ltd. develops AI-native industrial automation development platforms. Its flagship product, kVPAC IDE, is the world’s first production-grade IDE that deeply integrates LLM capabilities into industrial PLC programming workflows. Built on the VS Code architecture with a Rust high-performance engine, kVPAC IDE features an integrated MCP Server with 42+ AI tools, supporting natural language code generation, intelligent error diagnosis, and automated testing — boosting engineer productivity by over 60%.