Holiday Robotics

Scroll sideways for more →

Holiday Robotics (Korean: 홀리데이로보틱스) is a South Korean robotics company headquartered in the Gangnam district of Seoul that builds humanoid robots for industrial automation. Founded in March 2024 by Song Ki-young (송기영), the entrepreneur behind machine-vision startup SuaLab, the company makes FRIDAY, a wheeled humanoid that deliberately drops bipedal legs in favor of dexterous manipulation. Holiday Robotics raised a 17.5 billion KRW (approximately $13 million) seed round in August 2024, reported as the largest seed financing of any South Korean startup over the first eight months of that year, and on 13 July 2026 it announced a 155 billion KRW Series A that the company and Korean trade press described as the largest Series A ever raised by a South Korean startup. Its stated mission is "liberating people from physical labor with versatile robots."^{[1][2][3][4][17][21]}

Song is best known for co-founding SuaLab, a deep learning-based vision inspection company that was acquired by American industrial-vision leader Cognex in 2019 for approximately $195 million in what was at the time the largest foreign acquisition of a Korean technology startup. Holiday Robotics unveiled the first public prototype of FRIDAY on October 30, 2025, and by mid-2026 it was targeting commercial launch in the second half of that year at a target list price of roughly 100 million KRW per unit.^{[1][5][6][7][17][21]}

What is Holiday Robotics?

Holiday Robotics is a Seoul-based robotics startup, incorporated in March 2024, that develops general-purpose humanoid robots aimed first at industrial parts assembly and manufacturing, with a longer-term ambition to expand into service and household applications. Its single product to date is the wheeled humanoid FRIDAY. The company positions itself as a software-and-platform partner to factory system integrators rather than as a direct automation-services supplier, and its differentiation rests on two claims: that most of a humanoid's practical value in a factory comes from its hands rather than its legs, and that high-quality tactile sensing can be made roughly an order of magnitude cheaper than the prevailing market price.^[1][2][8]

As of July 2026 the company employed about 50 researchers and engineers alongside manufacturing and business-development staff, with recruits drawn from MIT, Seoul National University, KAIST and Samsung Electronics. It is headquartered in Seoul and describes itself as building a second engineering presence in the United States, where it was actively hiring in mid-2026.^[21][22][27]

Holiday Robotics describes its work as physical AI and organises it into four named systems: FRIDAY (the robot), Holiday Sim (the in-house simulator), Holiday Lab (the skill-learning platform), and OASys (the deployment and operations platform).^[18][27]

Who founded Holiday Robotics?

Holiday Robotics was founded and is led by Song Ki-young (송기영), a serial entrepreneur and engineer. The company romanizes his name as Kiyoung Song on its own team page, while Korean and English press coverage generally uses Song Ki-young.^[27]

What did Song Ki-young do before Holiday Robotics?

Song co-founded SuaLab in 2013 in Seoul as a developer of deep learning-based machine vision software for factory defect detection and quality inspection. Operating in the same broad domain as conventional rule-based machine vision but using convolutional neural networks rather than hand-engineered features, SuaLab built a customer base across Korean and East Asian manufacturing before being acquired by Cognex on October 18, 2019. Cognex paid approximately $195 million in cash for the company, a transaction widely reported in the Korean press as the largest M&A exit ever achieved by a Korean technology startup at that time.^[9][10]

At the time of the acquisition Song held a 25.54 percent stake in SuaLab. Stonebridge Ventures, which had backed the company since 2015 with roughly 13 billion KRW across multiple rounds, owned about 17 percent and reported an internal rate of return of approximately 105.79 percent on its position. After the deal closed Song joined Cognex as the head of its deep learning engineering division, where he served a four-year contractual employment period before leaving to start his next venture. He has cited the experience of running a deep-learning-driven product organization inside a global industrial vision company as a key influence on Holiday Robotics' decision to focus on physical manipulation rather than perception alone.^[9][11]

How was Holiday Robotics founded?

Holiday Robotics was incorporated in March 2024, immediately after Song completed his employment obligations at Cognex; Korean corporate-registry data compiled by startup databases gives an establishment date of 26 March 2024, and the company's own blog dates the founding to 2024. He recruited a founding team of AI software engineers and hardware controller specialists drawn from the Korean robotics ecosystem and from former colleagues at SuaLab and Cognex. In an early interview Song framed the new company as a continuation of his earlier work, saying that at SuaLab he "saw the potential of industrial AI" and now wanted to "challenge implementing that intelligence in physical space, the moving intelligence."^{[2][11][18][23][38]}

How much funding has Holiday Robotics raised?

Holiday Robotics has disclosed two priced rounds, totalling 172.5 billion KRW.

Scroll sideways for more →

The 2024 seed round

On August 27, 2024, just over four months after incorporation, Holiday Robotics announced that it had closed a seed round of 17.5 billion KRW (approximately $13 million, with USD conversions of $12.9 to $13.2 million reported depending on the exchange rate cited). Stonebridge Ventures led the round, with participation from SpringCamp, Atinum Investment, InterVest, and Hyundai Motor's open-innovation vehicle Zer01ne. The round set a record as the largest seed financing for a South Korean startup over the first eight months of 2024 and one of the largest disclosed seed rounds in Korean robotics history. Choi Dong-youl, chief investment officer at Stonebridge Ventures, justified the bet by noting that the founding team possessed "talent in both AI software and hardware controller development, which is critical for humanoid robots."^[4][5]

At the time of the announcement Holiday Robotics stated that it would deploy the proceeds primarily to hire specialists in robot control, reinforcement learning, large language models, and simulation engineering, and to establish an overseas robotics research center in the United States during the first half of 2025 to recruit international talent. The company has not published a separate announcement confirming that a US research center opened; as of July 2026 it described a "growing presence" in the United States and open roles there rather than an established facility.^{[5][6][18][27]}

The 2026 Series A

On 13 July 2026 Holiday Robotics announced a Series A of 155 billion KRW. The company said the round, which closed on 10 July, was the largest Series A raised by a South Korean startup and the largest single financing round raised by a domestic humanoid specialist. Dollar conversions differ across outlets and none is authoritative: Rocking Robots gave approximately $103 million, WOWTALE gave $103.4 million, and The Robot Report and Holiday Robotics' own blog both used $105 million. The won figure is the one the company itself stated, and the dollar equivalents are approximate conversions made by the publications at different exchange rates.^{[17][18][19][20][25]}

Existing backers Stonebridge Ventures, Atinum Investment, InterVest and SpringCamp all followed on. New investors were IMM Investment, SL Investment, KB Investment, BonAngels Venture Partners, Premier Partners, Ulmus Investment (얼머스인베스트먼트), SJ Investment Partners, Dasung Ventures, Atinum Capital Partners and Goodwater Capital, the last of these a US firm. Two Korean state-backed lenders, the Korea Development Bank and the Industrial Bank of Korea, also participated, which the company said gives it a wider base for future growth financing. No lead investor was identified in the announcement or in the coverage of it.^{[17][19][21][22][24][39]}

Holiday Robotics did not disclose a valuation. The financial outlet SmartToday derived an estimate from the company's share-issuance records, calculating that the 155 billion KRW was raised through 276,413 newly issued Class-2 redeemable convertible preferred shares at about 560,755 KRW each, which implies a post-money valuation of roughly 879.3 billion KRW and a pre-money valuation of about 724.3 billion KRW, with Series A investors holding 17.63 percent. By the same method SmartToday put the 2024 seed post-money valuation at about 77.5 billion KRW, making the implied increase roughly 11.4 times in 21 months. These figures are an outside calculation from filings, not a company disclosure, and the company has not confirmed them.^[23]

What will Holiday Robotics do with the Series A money?

The company said the proceeds go to three things: upgrading its mass-production system, expanding R&D headcount from about 50 to more than 100, and building manufacturing infrastructure capable of more than 10,000 units per year. It also restated the sequence of its long-term product plan, moving from manufacturing humanoids into service robots and then home robots. In the near term it said volume production should bring the unit price down to around 100 million KRW.^[17][21][22]

In its own announcement the company framed the round more narrowly than the Korean press did, saying the funding "sharpens our direction rather than changing it" and listing three uses: growing the research and engineering teams in the United States and Korea, strengthening manufacturing readiness, and continuing to develop the full-stack systems around FRIDAY.^[18][20]

What did Holiday Robotics do in 2025?

For its first eighteen months Holiday Robotics operated largely in stealth, restricting public communications to occasional founder interviews and partnership announcements. The company signed a strategic memorandum of understanding on August 1, 2025 with Endotlight (엔닷라이트), a Korean generative 3D AI startup led by Park Jin-young, to jointly produce 3D synthetic data for robot training. Under the agreement Endotlight supplies datasets generated by its Trinix synthetic data platform and proprietary 3D modeling engine, and Holiday Robotics applies them to perception and motion training for industrial scenarios. The two companies described the partnership as an effort to reduce the time and cost required to construct training data for physical AI systems.^[12]

On October 30, 2025 the company unveiled the first working FRIDAY prototype to media. The launch presentation emphasised that every core subsystem (the body structure, the actuators, the dexterous hands, and the magnetic tactile sensors) had been designed in-house from a clean sheet rather than assembled from off-the-shelf components, an approach Song described as building the robot "from the ground up." Holiday Robotics stated that it intended to demonstrate FRIDAY performing industrial tasks before the end of 2025 and to begin pilot deployments with manufacturing customers in 2026.^[1][13]

In September 2025 Song delivered a public lecture in which he described what he called the four open challenges that humanoid robots must overcome before reaching mass commercialization: slow learning (citing reports that Tesla Optimus had required engineers to repeat motions up to 20,000 times for data collection), the gap between simulation success and real-world reliability, safety risks of operating near humans (he noted that industrial robots cause an average of about 2.7 worker fatalities per year in Korea), and price. He singled out price as "the biggest stumbling block," arguing that while Chinese competitor Unitree advertises a $16,000 base unit, fully optioned configurations can add tens of thousands of dollars in tactile sensors and other accessories, with single-finger force sensors alone reportedly priced at around $14,000. He concluded that humanoids will not see mass adoption until their landed cost approaches that of an automobile.^[14]

Holiday Robotics was also named in 2025 among the humanoid makers working inside the K-Humanoid Alliance, the Korean government-backed consortium of ministries, research institutes and robot companies launched in April 2025. A Ministry of Trade, Industry and Energy briefing relayed by ZDNet Korea three weeks after the launch listed Holiday Robotics alongside Blue Robin and Robros as companies then in active discussion with major Korean manufacturers about factory deployment, and named it among the makers (Rainbow Robotics, A Robot, Holiday Robotics, Robotis and Robros) that would supply their in-house humanoids to research bodies including the Seoul National University AI Institute during 2025.^[40]

What did Holiday Robotics do in 2026?

2026 was the year Holiday Robotics moved from prototype disclosure to industrial partnerships, government programmes, and a large financing. The following events are all dated and independently reported, and the company indexes the underlying coverage in its own newsroom.^[37]

Scroll sideways for more →

The PowerAuto Robotics partnership

On 7 May 2026 Holiday Robotics signed a strategic partnership and joint development and supply agreement with PowerAuto Robotics, a Korean AI manufacturing-robot company that already supplies odd-form component insertion systems to Samsung Electronics production lines. The two companies said they would jointly develop humanoid automation for manufacturing sites and work toward a dark factory stack combining AI manufacturing robots, humanoids, process-control software and a real-time data platform. The specific target is manual odd-form insertion, the step after surface-mount technology in electronics assembly where connectors, relays and irregular components must be placed by hand and which has resisted conventional automation. PowerAuto Robotics chief executive Hwang Jang-sun called the agreement a starting point for extending to fully autonomous manufacturing processes, and Song said AI-driven autonomous manufacturing would determine global manufacturing competitiveness.^[34]

The national physical AI programme

On 9 June 2026 the Ministry of Science and ICT launched a Physical AI Leading Technology Development project with a budget of 34 billion KRW over two years, aimed at reducing Korean dependence on foreign simulators and platforms and producing a "Korean-style world model" for manufacturing and logistics. LG Electronics is the lead organisation, and the ten participants include Holiday Robotics, Robotis, Maum AI, Crowdworks, Alchera, KT, KAIST, Seoul National University and the Telecommunications Technology Association. Holiday Robotics' assigned role is to build a domestically developed simulator starting at the physics-engine level. Second Vice Minister Ryu Je-myung framed the programme as securing independent physical-AI infrastructure.^[35]

Government attention

On 8 July 2026 Park Hong-keun, Minister of Planning and Budget, held a humanoid roundtable in Gangnam, Seoul and visited Holiday Robotics to watch a live demonstration in which a robot collected and learned from data in real time, stacked components precisely, and moved heavy logistics loads. Park said the government would actively consider including a national manufacturing data library programme, run jointly by the Ministry of Science and ICT and the Ministry of Trade, Industry and Energy, in the 2027 budget.^[36]

What is the Friday robot?

FRIDAY is Holiday Robotics' first product and the platform on which the company plans to enter commercial service. Unlike many humanoid platforms that emphasise bipedal walking, FRIDAY is a wheeled mobile manipulator: it uses a differential-drive wheeled base instead of legs, a design choice the company argues delivers better stability, longer battery life, lower energy consumption, and predictable safety behaviour, all at substantially lower cost than a comparable biped. The product name reflects an internal roadmap in which weekday names denote distinct platforms: FRIDAY for industrial work, with later "Saturday" and "Sunday" variants planned for service and household applications respectively.^[1][8][27]

What are Friday's specifications?

Scroll sideways for more →

FRIDAY's body has 64 degrees of freedom in total, distributed across the head, arms, hands, waist and wheels, of which 40 sit in the two hands. Holiday Robotics presents that split as the clearest statement of its engineering priorities: "FRIDAY has 64 degrees of freedom, 40 of which are dedicated to its hands, a direct measure of where we put our engineering focus." The remainder is distributed across seven-DOF arms and a five-DOF torso. Press coverage from the October 2025 unveiling through the July 2026 funding announcement gives a total mass of 115 kg with 66 kg concentrated in the mobile base, a front-heavy configuration that gives the platform a low centre of gravity and high static stability; the company's own product page listed 125 kg as of late July 2026, and the discrepancy has not been explained publicly. Each arm is rated for a 5 kg payload at the wrist and up to 20 kg in proximal handling configurations. The base reaches a maximum speed of 1.9 m/s, equivalent to a brisk human walk.^{[1][3][8][17][19][20][23][26]}

The robot is designed for round-the-clock factory operation and uses a hot-swappable battery system so that battery exchanges can be performed during a shift without interrupting tasks; the company gives an average run time of four hours per battery. Onboard inference runs on an NVIDIA Jetson Orin module, with cloud connectivity for fleet learning and remote teleoperation as a fallback control mode. Holiday Robotics describes FRIDAY as offering both autonomous operation and a teleoperation mode, with control interfaces available through tablet, PC, voice, and gesture input.^[3][8][26]

Safety is presented as a hardware property rather than a software layer. The company lists an ISO 10218 compliant design, 360 degree obstacle detection using multiple cameras and lidars, a hardware emergency stop, lightweight low-inertia hands, and fully back-drivable hand and arm joints. Its stated position is that safe operation near people is a requirement rather than a feature.^[18][20][27]

What makes Friday's hands and tactile sensing distinctive?

The most distinctive subsystem on FRIDAY is its pair of in-house dexterous hands. Each hand has five fingers and 20 degrees of freedom, weighs under 600 grams including its integrated tactile sensors, and is rated for up to 5 kg of load per finger segment. The actuators support backdrivability across all joints, allowing the hand to comply softly with external forces rather than fighting them, a property useful for delicate assembly and human-safe interaction. At the October 2025 unveiling the hand was reported at about 500 grams.^{[1][6][15][26]}

In place of the proprietary multi-thousand-dollar tactile arrays sold by some humanoid suppliers, Holiday Robotics designed its own magnetic tactile sensor that costs approximately 30,000 KRW (around $21 to $23) per module to manufacture. The sensor uses a magnet embedded in a silicone surface; as external force deforms the silicone, the displacement of the magnet is read by a Hall-effect element underneath, which resolves both force and direction and can also infer torque. Reported figures for the sensor's resolution and update rate differ: the company's own product page gives 0.05 N sensitivity at a 250 Hz update rate, Humanoids Daily reported 0.05 N at 1,900 Hz from the unveiling, and ZDNet Korea reported a threshold of about 0.01 N (roughly 1 gram of contact force). Holiday Robotics covers approximately 80 percent of the hand's outer surface with these modules, including the entire palm and the inner surfaces of the fingers, rather than only the fingertips. Because the silicone wears with use, sensor pads are designed as replaceable modules, an approach Song has compared to trimming fingernails. The combination of full-surface tactile coverage and low per-sensor cost is central to the company's commercial pitch, since high-quality tactile feedback has historically been one of the most expensive line items on a dexterous hand.^{[1][8][16][26]}

Why did Holiday Robotics drop legs in favor of hands?

Song has repeatedly framed FRIDAY's design around the assertion that "most of the value comes from the hands" in an industrial setting. The company explicitly de-emphasises bipedal walking on the grounds that legs add cost, complexity, weight, and failure modes without delivering proportional value in a typical factory aisle, where a wheeled chassis can cover the relevant ground. Resources saved on legs are redirected to high-degree-of-freedom hands, dense tactile sensing, and a reinforcement-learning control stack. The company has also designed for safety predictability by selecting from a library of vetted behaviours rather than letting an end-to-end neural policy directly command motors, an approach discussed below. Humanoids Daily reported in November 2025 that the company had not abandoned legged locomotion outright and planned to test a bipedal version in December 2025; no result from that test has been published.^[1][14]

The company set out the underlying argument at length in a July 2026 essay titled "The Last 10 Centimeters." Its position is that factory automation has already solved the problem of moving objects between known states, and that what remains manual is the final moment of contact, when one part must meet another: a connector seating in a socket, a cable snapping into a guide, a gasket compressing evenly. Those steps fail not because the robot cannot see, the company argues, but because the information that determines success does not exist until something is touched. Holiday Robotics defines dexterity accordingly, not as fast fingers or complex gestures but as the ability to keep making progress when the world does not behave as expected, including recovering mid-motion rather than restarting. It also gives an economic version of the same argument: a hand is one of the few end effectors that does not have to be rebuilt when the task changes, which matters on lines that were laid out for people and whose product mix shifts. The essay is unusually candid about the company's stage, stating that FRIDAY has been through proof-of-concept work with manufacturing and logistics partners but has not been through sustained production use, and that until it has, "what is written here is an engineering position rather than a field result."^[28]

How does Friday's AI and software work?

What is Holiday Sim?

Holiday Sim is the company's proprietary simulation environment, designed specifically to close the so-called "sim-to-real" gap that has historically plagued reinforcement-learning approaches to manipulation. The simulator focuses on physically accurate modelling of soft-body contact, deformable materials, and friction, areas where general-purpose robotics simulators often perform poorly and which are central to dexterous manipulation tasks such as inserting flexible cables, picking deformable parts, or grasping objects with unknown surface properties. Policies trained in Holiday Sim are then transferred to physical FRIDAY hardware, where the company conducts iterative refinement.^[1][6]

By 2026 the company was describing Holiday Sim as a fully GPU-accelerated physics and policy generation engine, accelerated from contact dynamics through rendering, and positioning it for large-scale data generation as well as policy training. Alongside it the company disclosed Holiday World, a separate simulator built on learned neural networks that predict future states from physical understanding rather than from a hand-written physics engine, which it presents as a route to a smaller sim-to-real gap. Holiday Robotics' role in the national physical AI programme is to build this domestic simulator stack from the physics engine upward.^[26][35]

What are Holiday Lab and OASys?

In its Series A announcement the company described a four-part stack. FRIDAY is the robot; Holiday Sim reproduces the robots, their skills and their environment with enough fidelity that behaviours can be tested and broken before deployment; Holiday Lab converts demonstrations, simulation rollouts and real-world failures into reusable skills and then orchestrates and improves them; and OASys is the operational layer that deploys robots to a site and carries field experience back into the rest of the system. The company argues that owning the full stack matters less for control than for the direction information flows, since better hardware makes simulation more faithful, better simulation sharpens learning, and field failures are designed to propagate backward into the next version of the skill, the simulator and the robot.^[18][20][27]

How does the Vision-Language-Skill (VLS) architecture differ from VLA models?

In place of the vision-language-action (VLA) models that several competing humanoid developers have adopted, in which a single neural network maps visual and language inputs directly to motor commands, Holiday Robotics has built a proprietary Vision-Language-Skill (VLS) framework. In the VLS architecture, a perception and language stack interprets the scene and the task, but instead of emitting raw motor torques it selects from a library of pre-verified low-level control "skills" such as grasp, insert, push, pour, or place. Each skill is implemented and tested independently and exposes well-defined preconditions and post-conditions.^[1][3]

The company has described this as a "whitebox" approach in contrast to the largely opaque end-to-end policies emerging from VLA research. The advantages it claims are predictability and the ability to certify behaviour for industrial deployment, since each skill in the library can be inspected, tested, and bounded individually, and unsafe motions can be ruled out at the skill-selection layer rather than depending on emergent behaviour from a learned controller. The trade-off is that VLS will be less flexible than fully end-to-end policies for tasks that fall outside the existing skill library.^[1]

Holiday Robotics does build on vision-language models: its stated position is that VLMs have genuinely improved a robot's ability to interpret what needs to be done, but that interpreting a task and reliably performing it are different problems. Its specific objection to end-to-end mapping is diagnostic rather than philosophical: when perception maps straight to action, it is difficult to isolate why a particular motion failed, to fix it without retraining the whole task, or to reuse that behaviour elsewhere.^[18][20]

How does Friday learn new skills?

For learning new skills the company favours simulation-based reinforcement learning over imitation learning from human demonstrations, the approach used in many recent VLA systems. Song has compared the strategy explicitly to DeepMind's AlphaGo Zero, which mastered Go without human game records by playing against itself: the goal is to let the robot accumulate experience through thousands of trial-and-error iterations in a virtual environment that approximates reality, rather than relying on labour-intensive teleoperated demonstrations. The company has cited reported figures from competitors (including descriptions of engineers at Tesla repeating motions up to 20,000 times to collect data for the Optimus humanoid) as evidence that the demonstration-driven approach scales poorly. Holiday Robotics positions its self-play training pipeline as a way to minimise the data collection and engineering effort required to teach a humanoid a new task.^[6][11][14]

What is FlashSAC?

FlashSAC is a reinforcement-learning algorithm developed by Holiday Robotics researchers and academic collaborators, published as "FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control" (arXiv:2604.04539, April 2026) and released on GitHub under the MIT licence. On 22 July 2026 the company announced that the paper had received an Outstanding Paper Award at Robotics: Science and Systems (RSS) 2026, held 13 to 17 July 2026 in Sydney, Australia. Korean coverage described it as the first time a Korean team had won the conference's top paper award, noting that a KAIST team had previously received the systems paper award in 2023, and made a point of the fact that the work was led from a startup rather than a university or national laboratory.^{[29][30][31][32][33]}

The technical problem FlashSAC addresses is the standard trade-off in robot learning between on-policy methods such as proximal policy optimization, which are stable but evaluate policies poorly in high-dimensional state and action spaces because their data is narrowly distributed, and off-policy methods, which learn from a broader distribution but converge slowly and unstably because fitting a value function over diverse data requires many gradient updates and lets critic errors accumulate through bootstrapping. FlashSAC is built on Soft Actor-Critic and takes the opposite approach suggested by supervised-learning scaling laws: it sharply reduces the number of gradient updates while compensating with larger models and higher data throughput, and explicitly bounds weight, feature and gradient norms to keep training stable at that larger scale.^[29]

The paper reports that across more than 60 tasks in 10 simulators FlashSAC outperforms PPO and strong off-policy baselines on both final performance and training efficiency, with the largest gains on high-dimensional tasks such as dexterous manipulation. In sim-to-real humanoid locomotion it reduces training time from hours to minutes. The authors verified the transfer on physical hardware, reporting walking, turning and stair traversal from policies trained in simulation. Holiday Robotics says the algorithm is applied in FRIDAY's development rather than remaining a research result, and that shortening training time compresses the loop between designing a new motion and running it on a real robot. The paper's author list combines Holiday Robotics staff (including lead authors Donghu Kim and Hojoon Lee, plus I Made Aswin Nahrendra and Sehee Min) with academics from KAIST, TU Darmstadt, KTH Royal Institute of Technology, KRAFTON and the German Research Center for Artificial Intelligence.^[29][30][32]

What is Holiday Robotics' business model and roadmap?

Holiday Robotics has positioned itself as a software-and-platform partner to existing system integrators rather than a direct supplier of factory automation services. Under this business-to-business model the company supplies FRIDAY units and the underlying control software, while integrator partners handle deployment, line-side integration, and customer support. Song has said the company is currently conducting proofs of concept with major Korean manufacturers, although specific customer names had not been disclosed publicly as of the company's late-2025 announcements.^[11][14]

As of July 2026 the company said it was running confidential proof-of-concept programmes with major domestic and international companies in automotive, semiconductor and logistics, sectors it selects for their high proportion of repetitive work. It has not disclosed the customers, the number of robots involved, or the specific applications, a gap that Rocking Robots noted in its coverage: the production targets below are planned capacity rather than confirmed deliveries, and no customer orders, deployment results or performance data from the trials have been published.^[17][20][25]

When will Friday go into production and how much will it cost?

The company's publicly stated production roadmap, as restated with the Series A in July 2026, is:

Scroll sideways for more →

The company has stated a target list price of approximately 100 million KRW per unit, a figure it justifies by reference to the annualised cost of a manufacturing worker. Dollar equivalents given for that price have moved with the exchange rate and with the outlet: Humanoids Daily used roughly $70,000 in late 2025, while WOWTALE and Rocking Robots used about $66,700 and $67,000 respectively in July 2026. Holiday Robotics has not said whether the price would include software, installation, integration or ongoing support. Song has been explicit that he views Holiday Robotics as a long-horizon, scale-driven business and has publicly favoured an eventual initial public offering rather than another early acquisition exit, arguing that the addressable market for humanoid labour is potentially larger than the combined automotive and smartphone industries. With the Series A he added a longer-range ambition, saying the company aims "to grow into one of the world's top 10 robotics companies."^{[1][11][8][16][17][19][25]}

How was Holiday Robotics received?

Holiday Robotics' August 2024 seed round drew attention in the Korean tech press for its size, especially against the backdrop of a broader funding slowdown for Korean startups in 2024. KED Global highlighted the round as one of several outliers showing that AI and robotics ventures continued to attract capital despite the wider funding drought, and Stonebridge Ventures' Choi Dong-youl publicly described the team's combined AI-software and hardware-controller expertise as the principal investment thesis.^[4]

The October 2025 unveiling of FRIDAY was covered by both Korean technology outlets and English-language humanoid trackers including Humanoids Daily, which characterised the design as a deliberate counter-position to bipedal-first humanoid programmes by Tesla, Figure AI, Agility Robotics, and Chinese rivals such as Unitree Robotics. Coverage focused in particular on the claim that Holiday Robotics had reduced the per-sensor cost of high-resolution tactile feedback to roughly $20 through its custom magnetic-Hall-effect design, an order-of-magnitude reduction from prevailing market prices for comparable sensors. Analysts cited by Korean outlets noted that price-aggressive industrial robot entrants from China, particularly Unitree, would be the company's most direct commercial competitors and that the wheeled-base, dexterous-hand design represented a coherent strategy for differentiating on cost and reliability rather than on athletic performance.^[1][14]

The 2026 Series A shifted the framing from technical curiosity to financial outlier. Korean coverage led on the record: SmartToday noted that the implied post-money valuation of roughly 879.3 billion KRW put a Series A company at the level Rebellions and Xcena had reached at Series B, and about 2.9 times the valuation of Bos Semiconductor, another Korean deep-tech company that had just raised its own Series A. The same analysis identified the risk directly, observing that Holiday Robotics secured a valuation in the 800 billion KRW range ahead of any meaningful production or revenue, so converting proofs of concept into paid supply contracts is the condition for holding that valuation. English trade coverage struck a similar note: Rocking Robots wrote that establishing a 100-unit production system would itself be a change in the company's development phase, requiring it to solve component sourcing, assembly, quality control, maintenance and integration into existing factory processes, and that the next test is whether FRIDAY performs reliably enough to justify its purchase and integration costs.^[23][25]

Within the humanoid robot market, Holiday Robotics' round was one of several large 2026 financings that went to wheeled rather than legged platforms, a pattern that industry coverage attributed to the lower cost, mechanical simplicity and easier safety certification of a wheeled base. The RSS 2026 award added a research dimension to a company that had previously been discussed mainly in terms of hardware pricing.^[20][32]

See also

• FRIDAY
• Humanoid robots
• Humanoid robot market
• Physical AI
• Tesla Optimus
• Unitree
• Agility Robotics
• Figure AI
• Vision-language-action model
• Reinforcement learning
• Industrial robot

References

1. ^"Holiday Robotics Unveils 'Friday': A Wheeled Humanoid That Prioritizes Hands Over Legs." Humanoids Daily.
2. ^"Holiday Robotics company profile." Aparobot.
3. ^"Friday by Holiday Robotics." Humanoid.guide.
4. ^"Korea's AI, robot startups shine despite funding drought." KED Global.
5. ^"Holiday Robotics Secures KRW 17.5 Billion in Seed Funding for Humanoid Robot Development." WOWTALE.
6. ^"홀리데이로보틱스, 1300만 달러 시드 투자 유치." Robot Newspaper (로봇신문).
7. ^"Holiday Robotics Secures $13M for Humanoid Robot Development." WOWTALE.
8. ^"Friday by Holiday Robotics: Price, Details, Review." Origin of Bots.
9. ^"이번엔 휴머노이드로 승부…2300억 엑시트 창업가의 두번째 도전." Unicorn Factory.
10. ^"Nasdaq-listed Cognex acquires S. Korean AI startup Sualab." The Korea Herald / The Investor.
11. ^"창업 1년 반 만에…홀리데이로보틱스, 첫 휴머노이드 공개." Unicorn Factory.
12. ^"엔닷라이트, 홀리데이로보틱스와 전략적 업무협약 체결." Venture Square.
13. ^"Why Cognex Acquired SUALAB: Deep Learning." Cognex Blog.
14. ^"송기영 홀리데이로보틱스 대표 '4대 과제 해결 못하면 휴머노이드 상용화 안 된다'." Hankyung.
15. ^"Friday robot details." Aparobot.
16. ^"3만원 센서로 만든 휴머노이드 손…내년 양산 개시." ZDNet Korea.
17. ^"Humanoid Developer Holiday Robotics Raises 155 Billion Won." Seoul Economic Daily (English edition), July 13, 2026.
18. ^"$105M Series A: Dexterous Humanoids for Real Industrial Work." Holiday Robotics blog, July 13, 2026.
19. ^"Holiday Robotics Raises $103.4 Million Series A to Scale Humanoid Robot Manufacturing." WOWTALE, July 13, 2026.
20. ^"Holiday Robotics raises $105M for its FRIDAY wheeled humanoid." The Robot Report, July 23, 2026.
21. ^"휴머노이드 홀리데이로보틱스 1550억 유치…국내 시리즈A 역대 최대." Unicorn Factory (머니투데이), July 13, 2026.
22. ^"휴머노이드 스타트업 홀리데이로보틱스, 1550억원 투자 유치." Robot Newspaper (로봇신문), July 13, 2026.
23. ^["[홀리데이로보틱스 디코드] 시리즈 A서 기업가치 8800억…리벨리온보다 빠르다." SmartToday, July 13, 2026.](smarttoday.co.kr/...109446)
24. ^"Holiday Robotics attracts 155 billion won Series A investment." Maeil Business Newspaper (English edition), July 13, 2026.
25. ^"Holiday Robotics Raises $103 Million to Move Humanoid into Industrial Production." Rocking Robots, July 15, 2026.
26. ^"FRIDAY product page." Holiday Robotics, accessed July 31, 2026.
27. ^"About Us." Holiday Robotics, accessed July 31, 2026.
28. ^"The Last 10 Centimeters." Holiday Robotics blog, July 30, 2026.
29. ^Kim, Donghu; Lee, Youngdo; et al. "FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control." arXiv:2604.04539, April 2026.
30. ^"Holiday-Robot/FlashSAC." GitHub repository, accessed July 31, 2026.
31. ^"Holiday Robotics Wins Best Paper Award at World's Top Robotics Conference." Seoul Economic Daily (English edition), July 22, 2026.
32. ^"학습 시간은 '수 시간에서 수 분으로'... 홀리데이로보틱스, 세계 최고 로보틱스 학회서 한국 첫 최우수논문상." Venture Square, July 22, 2026.
33. ^"Robotics: Science and Systems 2026." RSS Foundation, Sydney, Australia, July 13 to 17, 2026.
34. ^"파워오토로보틱스-홀리데이로보틱스, AI 기반 자율 제조 혁신 협약." Robot Newspaper (로봇신문), May 7, 2026.
35. ^"엔비디아 의존도 낮춘다…과기부, 피지컬 AI 국산화 시동." Bloter, June 9, 2026.
36. ^["[포토] 박홍근 장관, 휴머노이드 로봇 기업 홀리데이로보틱스 현장방문." Hankyung, July 8, 2026.](hankyung.com/...202607083686i)
37. ^"Newsroom." Holiday Robotics, accessed July 31, 2026.
38. ^"홀리데이로보틱스 기업정보." NextUnicorn, accessed July 31, 2026.
39. ^"홀리데이로보틱스, 1,550억 규모 시리즈 A 투자 유치…역대 최대 규모 기록." beSUCCESS, July 13, 2026.
40. ^"K-휴머노이드 연합, 출범 3주 만에 협약 4건 성과." ZDNet Korea, May 1, 2025.