Deep Capital Leads Investment, This Micro-Joint Module Enterprise Completes Tens of Millions in Financing | 36Kr Exclusive 深创投领投，这家微型关节模组企业完成数千万融资丨36Kr首发

Investors’ press releases are filled with buzzwords like "pioneers of uncharted territory," "disruptive thinking," and "core bottleneck." Calm down, strip away the fancy rhetoric, and what’s the essence we see? It’s a team founded last August—less than a year old—attempting to tackle an industry-recognized, long-standing problem using a somewhat "non-mainstream" combination of technologies: axial flux motors, PCB windings, and cycloidal reducers. The goal is to cram a sufficiently powerful, precise, and affordable drive unit into an extremely small space. Founder Zhang Yang puts it plainly, with the clarity of a latecomer: "We entered the market late, so we had to find an unconventional path." In plain terms, this means a head-on clash with overseas players who’ve深耕多年 in radial flux motors or traditional coreless motor designs is hopeless—they can only launch a "sneak attack" via an alternative technological route.

Their chosen technical path does have its logical elegance. Axial flux motors inherently offer higher torque density potential, but miniaturization is universally recognized as a "hellish difficulty level." PCB windings replace manual coil winding with semiconductor manufacturing processes, solving the nightmare of precision and consistency in micro motors. Cycloidal reducers aim to strike a balance between cost and reliability. Combining these three on a millimeter scale indeed shows a certain "Frankenstein-like" innovative courage. But the critical question remains: Is this a proven, viable highway, or merely an elegant concept from the lab? The founder’s emphasis that "miniaturization is our biggest moat" deserves a question mark. Once a technical route proves feasible, followers will imitate and optimize at breakneck speed. They claim to be the "only" company integrating these three technologies and miniaturizing them, but how long can that time window last?

This exposes the most paradoxical aspect of current robotics core-component startups: Everyone talks about "solving choke-point problems," but many solutions themselves seem to be choked by the bottlenecks of "mass-production consistency" and "long-term reliability." Zhang Yang candidly criticized industry claims about dexterous hand lifespan in an interview: "2 million cycles only add up to just over 200 hours—even household appliances don’t meet such low standards." This sharp remark directly pricks the bubble under capital-driven narratives. Something that can’t even perform better than a light switch is still far from entering factories and homes; what stands in the way isn’t funding but tens of thousands of testing hours and countless process-detail pitfalls. Fingertip Intelligence claims to be advancing reliability testing and says its "advantages are quite obvious," but this advantage currently exists only on PPT slides and in lab data. For a startup, surviving the rigorous delivery tests of key clients (like the two that have "finalized strategic development partnerships") and turning planned production capacity of hundreds of thousands or even millions of units into a steady flow of qualified products is far more important than any tech pitch.

Looking at their business model, Zhang Yang repeatedly emphasizes: "Our positioning is clear—we’re the 'shovel seller'," and "We won’t venture downstream." This is a very smart and pragmatic positioning. When even whole-unit (dexterous hand) manufacturers are struggling with cost, reliability, and integration, a component maker rushing in would likely be squeezed from both sides—failing at making components while also poaching customers’ business. They understand keenly, "It’s not that we can’t do it, but there’s no need. Besides, many clients worry about us moving downstream." This sense of boundaries is crucial in a stage where the ecosystem is still fluid. However, this also means they must sharpen their "shovel" enough—making it sufficiently锋利, unique, and attractively priced—so that the gold rushers have no choice but to buy. Their imagination is indeed anchored in the larger market for "high-performance micro motors," covering medical, consumer electronics, and other fields. Yet the barriers in these areas (like medical certifications, extreme cost control in consumer electronics) are no less formidable than in robotics—they’re battles on another dimension. Whether entering via the hot concept of "dexterous hands" can truly radiate into these more traditional markets that depend on channels and accumulated experience is a longer-term test.

Ultimately, this funding round for Fingertip Intelligence, and the wave of "hardcore" component companies it represents, marks a critical chapter in the evolution of China’s robotics industry from "integrated innovation" to "source-component innovation." This chapter is being written in fits and starts, filled with technological gambles, the struggle of production hell, and commercial-strategy trade-offs. What institutions like Shenzhen Capital Group are betting on is the urgent demand for core components amid the embodied intelligence wave, and this team’s possibility of applying a semiconductor mindset to motor production—a "dimensionality reduction" approach. But beneath the wave lies only silt and sand. The real test isn’t securing funding, but whether these funds can first power them to the finish line in the dry marathon of reliability testing, help them walk the tightrope of cost control, and ultimately transform that "dexterous hand" from a showpiece on a demo stage into a reliable tool on a production line. Until then, all titles like "core engine" or "disruptive" remain mere hopeful aspirations.

数千万元Pre-A轮，深创投领投，一家叫“指尖智擎”的初创公司。这事放在两年前的机器人赛道，可能激不起半点水花，但在2025年这个被资本和叙事双重炙烤的“具身智能元年”，它像一滴水落入热油，炸开的讯号值得咀嚼。它炸开的不是一个宏大的故事，而是一个具体、琐碎、却卡住所有人喉咙的难题：机器人那双灵巧的手，什么时候才能真正“灵巧”且“可靠”地干活？

投资方的通稿里充斥着“无人区开拓者”、“颠覆性思路”、“核心瓶颈”这类词汇。冷静一下，把这些华丽的辞藻剥掉，我们看到的本质是什么？是一个成立于去年8月、不到一年的团队，试图用一套看起来有点“非主流”的技术组合——轴向磁通电机、PCB绕组、摆线减速器——去解决一个行业公认的老大难问题：在极小的空间内，塞进足够强大、精密且便宜的驱动单元。创始人张杨的话很实在，也透着一股后发者的清醒：“切入市场较晚，所以必须另辟蹊径”。这话说白了，就是正面硬刚海外那些深耕多年的径向磁通电机或传统空心杯电机方案，希望渺茫，只能在技术路线上“偷袭”。

他们选择的技术路线确实有其逻辑上的精妙。轴向磁通电机天生有更高的扭矩密度潜力，但微型化是公认地狱级副本；PCB绕组用半导体工艺取代手工绕线，解决了微型电机精度和一致性的噩梦；摆线减速器则试图在成本与可靠性之间找平衡。把这三者在毫米尺度上捏合在一起，确实有点“缝合怪”式的创新勇气。但问题在于，这究竟是一条经过验证的康庄大道，还是仅仅是一个实验室里的精巧构想？创始人强调的“微型化是最大护城河”，这话得打个问号。技术路线一旦被证明可行，追赶者的模仿和优化会来得极快。他们自称是“唯一”把这三项技术结合并做微型化的公司，但时间窗口能有多久？

这里就暴露了当前机器人核心部件创业最吊诡的一点：大家都在谈“解决卡脖子问题”，但很多解决方案本身，似乎也卡在了“量产一致性”和“长期可靠性”的脖子上。张杨在采访里吐槽行业宣称的灵巧手寿命，“200万次也就200多个小时，连家用电器的标准都达不到”，这话相当犀利，直接戳破了资本叙事下的泡沫。一个连开灯关灯都不如的东西，距离走进工厂和家庭，中间隔着的不是融资金额，而是数以万计的测试小时和无数工艺细节的魔鬼。指尖智擎宣称已在推进可靠性验证，并称“优势比较明显”，但这份优势目前还只停留在PPT和实验室数据阶段。对于一家初创公司，如何在头部客户（比如那两家“敲定战略开发合作”的）严苛的交付测试中活下来，并把产能从规划中的十万台、二十万台，实实在在地变成稳定流出的合格产品，这比任何技术演讲都更重要。

再看他们的商业模式。张杨多次强调，“我们定位很清晰，就是做‘卖铲子的人’”、“不去干下游的事”。这是一个非常聪明且务实的定位。在整机（灵巧手）厂商自己都在为成本、可靠性和集成焦头烂额的时候，部件商若贸然冲进去，大概率是两头受夹，既做不好部件，又抢了客户的饭碗。他们深知“不是不能做，而是没必要。况且很多客户担心我们做下游”，这种边界感在生态位尚未稳定的阶段至关重要。但这也意味着，他们必须把“铲子”磨得足够锋利、足够独特，且价格诱人，才能让淘金者不得不买。他们的想象空间确实被锚定在了“微型高性能电机”这个更大的市场，覆盖医疗、消费电子等。然而，这些领域的壁垒（如医疗认证、消费电子的极致成本控制）丝毫不低于机器人，甚至是另一个维度的战争。从“灵巧手”这个火爆概念切入，能否真正辐射到这些更传统、更看渠道和积累的市场，是更长期的考验。

归根结底，指尖智擎的这轮融资，以及它所代表的一批“硬核”部件公司，正在演绎中国机器人产业从“集成创新”走向“源头部件创新”的一个关键章节。这章写得磕磕绊绊，充满了技术路径的赌博、量产地狱的挣扎和商业策略的权衡。深创投等机构押注的，是具身智能浪潮下对核心零部件的刚需，以及这个团队用半导体思维做电机的“降维”可能性。但浪潮之下，尽是泥沙。真正的考验不在于拿到钱，而在于能否用这些钱，在可靠性测试的枯燥长跑中率先撞线，在成本控制的钢丝上走出平衡，并最终让那双“灵巧手”，从演示台上的炫技，变成流水线上可靠劳作的工具。在此之前，一切“核心引擎”和“颠覆性”的称号，都只是美好的期许。