Heart Capital Makes Consecutive Investment in Kunlun Robotics, Accelerating Commercialization of Embodied Intelligence | HeartStar PORTFOLIO

Kunlun Robotics recently announced the completion of a three-round series, with cumulative funding reaching several billion RMB less than three months after incorporation — setting a new record for the fastest fundraising by a domestic embodied-intelligence startup going from zero to one in 2026.

Heart Capital doubled down across two consecutive rounds in this series, alongside Gaorong Ventures, Hillhouse, CASSTAR, Eastern Bell Capital, Huaye Capital, Sinovation Ventures, and C&D Capital, the investment arm of C&D Group.

Liu Yanchen, Partner at Heart Capital, said: "Congratulations to Kunlun Robotics on completing this three-round series. What Heart Capital values is the team's rare compound skill set — hands-on engineering experience taking autonomous driving from R&D to mass production, combined with the commercial ability to run and manage large organizations. This 'technology + engineering + commerce' combination is uncommon in the embodied-intelligence track. We believe it's exactly the key capability needed to navigate technological uncertainty and achieve true industrial deployment."

Three Rounds Closed at Breakneck Speed

Kunlun Assembles an "Ace Founding Team"

Three funding rounds in under 90 days, setting this year's record for zero-to-one startups — what drove investors' extraordinary enthusiasm for Kunlun Robotics was its assembly of a top-tier "ace founding team."

Founder and CEO Ren Geng is a versatile leader whose résumé combines leading Alibaba Cloud to hundred-billion-RMB revenue as its top commercial executive, serving as an overseas country CEO during Huawei's globalization push, and running a conglomerate of over 50,000 employees. At Huawei, Ren was among the youngest country CEOs, overseeing all three business groups — carrier, enterprise, and consumer — in his host country, spanning solutions, sales, delivery, procurement, supply chain, and government and public relations. After joining Alibaba, he became one of the youngest of the post-80s generation to reach VP rank, holding core roles including President of Alibaba Cloud China, President of Video Cloud & Edge Cloud R&D, and Chief of Staff for Alibaba Cloud. Under his leadership, Alibaba Cloud China maintained market-share leadership, peaking at 42.1% in 2020. Alibaba insiders described him as a field commander who could "take on the hard jobs, deliver results, and win battles" — a practitioner skilled end-to-end across strategy, product, commerce, operations, and organization. After leaving Alibaba, Ren became President of ENN Group, driving the 50,000-plus-person conglomerate's expansion into industry large models, digital agents, and embodied intelligence.

Ren Geng, Founder and CEO of Kunlun Robotics

Co-founder Lang Xianpeng is one of the very few AI technology leaders in embodied intelligence who has "taken autonomous driving from zero through one to one hundred," bringing both cutting-edge research and mass-production experience. Previously, he served as President of Autonomous Driving at Li Auto and was the "first employee" of its intelligent driving division, personally building Li Auto's complete self-driving R&D system from scratch, assembling the core technical team, and establishing the industry-benchmark data flywheel that underpinned Li Auto's rise to the top tier. He led "Project Weicheng" to push for full in-house development of Li Auto's intelligent driving, delivering highway NOA, urban NOA, and mapless NOA to production. In 2024, he led his team to complete the industry's first end-to-end plus VLM solution in under 100 days; in 2025, he launched and mass-produced the VLA model in autonomous driving, cementing Li Auto's place in the first tier.

Lang Xianpeng, Co-founder of Kunlun Robotics

A versatile leader with first-hand experience running tech giants at the top level, paired with a technical mind who understands both frontier research and mass production — this founder combination is extraordinarily scarce in the embodied-intelligence space.

Beyond the founders, Kunlun has assembled a "most scarce in the industry" team with "proven success." The team includes algorithm veterans and R&D leads who lived through Li Auto autonomous driving's entire zero-to-one-to-one-hundred journey with deep AI deployment experience; system builders who led hardcore tech teams at Huawei and operators who shipped Huawei hardware products by the millions; and hybrid R&D-plus-commercial talent from Alibaba Cloud with battle-tested experience cracking large state-owned enterprise markets and landing industry solutions. From founders to core team, Kunlun established differentiated competitive advantages from day one:

At the company level, a tech-giant leader as the number-one operator keeps the enterprise on the right strategic course. On technology, top-tier AI deployment practice and elite hardware R&D systems ensure cutting-edge research makes it out of the lab. On commercialization, genes proven at hundred-billion-plus scale support full-cycle operations from sales to delivery, closing the loop from technology to business.

Committed to Hardware-Software Integration

Building a Mass-Producible Full-Stack Technical System

As industry investment sentiment shifts, capital markets have moved on from concept-driven, demo-grade project hype to focus on the trinity of hardcore physical platforms, embodied-AI brains, and commercial mass-production capability. From inception, Kunlun anchored on engineering for mass production and scenario-based commercial deployment, committing to full-stack self-development of both AI brain and hardware platform to build a fully autonomous and controllable technical system.

Ren Geng believes that AI, as the core of intelligence, determines the ceiling of embodied intelligence; hardware, as the carrier of intelligence, determines the floor; the two are complementary and indispensable. Sustained AI innovation capability and hardware platform R&D constitute the technical moat of an embodied-intelligence company. But Ren also emphasizes that beyond these two technical systems, truly achieving mass production and closing the commercial loop requires building the commercial system from opportunity to cash collection, operating the supply-chain ecosystem, and establishing delivery and service systems — these together create the true "mass-production acceleration" of embodied intelligence.

On specific technical approach, Kunlun follows a hardware-software integrated path targeting high-quality mass production, building full-stack self-developed core capabilities while simultaneously deepening both "brain" and hardware platform R&D.

Addressing industry-wide pain points including lack of physical causal reasoning, poor scenario generalization, and "black box" model decision-making, Kunlun introduces physical causal modeling on top of the mainstream Vision-Language-Action Model (VLA) framework, constructing a dual-system robotic intelligence architecture centered on the Kunlun World Model (KWM) and making original technical contributions across physical reasoning, data efficiency, and cross-embodiment transfer.

On data, Kunlun is the first to upgrade from "data flywheel" to "data compilation," rejecting the industry habit of crude data volume scaling. Through structured processing, task explicitization, and failure-sample accumulation, raw data is transformed into core assets for causal learning, forming a self-growing, sustainably iterative data production line that builds long-term technical moats.

Facing the industry's lack of evaluation frameworks for true model generalization, Kunlun has also built the industry's first counterfactual evaluation system, using real physical outcomes as the judgment standard to precisely distinguish genuine generalization capability among models with equivalent fitting accuracy.

From model to data, layered with the counterfactual evaluation system and "one brain, multiple forms" cross-embodiment transfer technology, Kunlun's "brain" solution has a clear objective: high-efficiency model iteration, low-cost scenario and cross-embodiment generalization, breaking through the industry bottleneck where lab technology struggles to scale — everything oriented toward engineering for mass production.

Kunlun has built an integrated full-stack hardware technical system spanning platform, motion control, and integration testing. In development logic, it adopts a software-defined-hardware forward development model that fundamentally solves the chronic industry problem of hardware-software disconnection and mismatch between motion execution and AI planning; underlying motion-control core algorithms are fully self-developed, ensuring high-precision, low-jitter, high-compliance, strong anti-interference operation; in platform architecture design, heterogeneous platforms employ modular architecture with unified electrical and communication interfaces, lowering the hardware barrier for model cross-embodiment generalization.

Going forward, Kunlun Robotics will stay anchored on the long-term track of general-purpose embodied intelligence, continuing to refine its full-stack self-developed core technologies, expanding the boundaries of robotic mass production and deployment, building complete industrial synergy networks alongside supply-chain partners and capital, and leading the embodied-intelligence industry toward scaled, commercialized transformation.

---

Founded in 2022, Heart Capital is a venture capital fund focused on investing in China's early-stage technology startups.

Heart Capital's team is primarily composed of founding partners and core investors from Lightspeed China, alongside seasoned investors from industry. The team's past investments include MetaX (688802.SH), Xpeng Motors (NYSE: XPEV, 09868.HK), Full Truck Alliance (NYSE: YMM), Sunmi Technology (06810.HK), RoboSense (02948.HK), Ambiq Micro (NYSE: AMBQ), Hanshow Technology (301275.SZ), FinVolution (NYSE: FINV), Qimengdao Group (NASDAQ: HERE), as well as LandSpace, MinoSpace, Baichuan, Manbang Cold Chain, World Logistics, Fan Deng Reading, and Lanhu.

Rooted in China with a global outlook, Heart Capital is committed to early-stage陪伴 and supporting entrepreneurial teams with the potential to become world-class companies in China's technology sector. Heart Capital advocates the value of "heart," believing that technology can serve as a bridge connecting minds. Heart Capital looks forward to accompanying more young Chinese entrepreneurs onto the world stage.