A new layer of the AI economy is quietly forming—not around the models themselves, but around the infrastructure that turns those models into usable systems inside real organizations. Dify, an open-source platform designed to help teams build, deploy, and operate production-grade AI applications, has raised $30 million in Series Pre-A funding at a $180 million valuation. The round was led by HSG, with participation from GL Ventures, Alt-Alpha Capital (a spin-out from Bessemer Venture Partners), 5Y Capital, Mizuho Leaguer Investment, and NYX Ventures.
The company positions itself at a critical transition point in the AI market. For the past two years, much of the industry has been focused on experimentation—building prototypes, testing models, and exploring possibilities. Now organizations are beginning to move from demos to production systems. That shift changes everything. Instead of simple prompts or one-off chatbots, companies need structured workflows that integrate AI into business operations. Document pipelines, internal copilots, customer support automation, invoice auditing, compliance checks—these are systems that must run reliably, connect to databases and internal tools, and be maintainable over time.
Dify was built to serve exactly that role. The platform provides a visual workflow builder combined with the engineering infrastructure required for production environments. Teams can design agent-based workflows while managing prompts, connecting tools, retrieving knowledge from internal data sources, debugging system behavior, and deploying APIs. In effect, Dify sits between foundation models and enterprise systems, acting as the orchestration layer that translates raw AI capability into structured operational workflows.
According to founder and CEO Luyu Zhang, the core idea behind the platform is straightforward: organizations should be able to transform their own domain knowledge and internal processes into AI applications without having to rebuild infrastructure every time. That philosophy reflects a broader industry shift toward what many now call “agentic workflows”—systems in which AI models perform sequences of tasks across tools, data sources, and decision points rather than simply answering questions.
Since its launch in 2023, Dify has grown quickly in the developer ecosystem. The project ranks among the most-starred open-source repositories on GitHub and is now running on more than 1.4 million machines worldwide. More than 2,000 teams and around 280 enterprises are using commercial versions of the platform. Among them are organizations such as Maersk, ETS, Anker Innovations, and Novartis, all experimenting with ways to embed AI directly into operational processes.
For companies like Maersk, which manages complex logistics workflows across global shipping networks, the appeal is obvious. Instead of building AI infrastructure from scratch for every new automation project, teams can construct reusable workflows that connect AI models to real operational systems. According to Mark Sear, Director of AI Solutions Engineering at Maersk, the platform made it possible to turn experimental prototypes into maintainable production systems.
Investors see the opportunity as part of a much larger structural shift. The AI application layer—software built on top of large language models and other AI systems—is still in its early stages, but it is accelerating rapidly. Platforms that standardize development workflows and provide the operational backbone for AI deployment could become long-term infrastructure companies. Kui Zhou, partner at HSG, described the space as one where standardized development platforms are likely to capture durable value as AI becomes embedded in everyday business operations.
The newly raised funding will be used to expand Dify’s 2026 roadmap. Plans include deeper integration of advanced agent capabilities, improvements to observability and reliability, and the creation of a dedicated enterprise product team focused on compliance and performance. The company also intends to expand its builder ecosystem and continue lowering the barrier for teams looking to deploy AI systems across departments.
Viewed from a broader perspective, Dify’s rise reflects a pattern that often repeats in technology cycles. Early breakthroughs tend to focus on raw capability—in this case, powerful foundation models. The next phase centers on the tools that allow organizations to operationalize that capability. Platforms like Dify are emerging in the space between experimentation and real-world deployment, where AI systems must behave less like demos and more like infrastructure.
If that transition continues at its current pace, the companies enabling reliable, observable, and maintainable AI workflows may end up shaping how the next generation of software is actually built.
Leave a Reply