{"id":53722,"date":"2026-05-26T12:24:45","date_gmt":"2026-05-26T16:24:45","guid":{"rendered":"https:\/\/overcentral.com\/en\/human-archive-raises-8-2-million-to-train-robots-using-indian-gig-workers\/"},"modified":"2026-05-26T12:26:21","modified_gmt":"2026-05-26T16:26:21","slug":"human-archive-funding-robot-training","status":"publish","type":"post","link":"https:\/\/overcentral.com\/en\/human-archive-funding-robot-training\/","title":{"rendered":"Human Archive Raises $8.2 Million to Train Robots Using Indian Gig Workers"},"content":{"rendered":"

In a funding round that underscores the escalating demand for physical AI training data, Silicon Valley-based Human Archive has announced it raised $8.2 million<\/a> from a consortium that includes Wing Venture Capital, NVP Capital, Y Combinator, and angel investors affiliated with OpenAI<\/a>, Nvidia<\/a>, Google<\/a>, Mercor, AfterQuery, BAIR, SAIL, and Meta. The startup is building a pipeline of egocentric video and multi-sensor data by equipping gig workers in India’s booming home services sector with camera-mounted caps and specialized wearable devices, capturing first-person footage of everyday tasks that could one day teach robots how to clean, cook, and navigate human environments.<\/h2>\n

Founded by two University of California, Berkeley students and two Stanford University students \u2014 Samay Mani, Rushil Agarwal, Shloke Patel, and CEO Raj Patel, the latter two being cousins \u2014 Human Archive is a direct bet on where the artificial intelligence industry is heading. As robotics labs and frontier AI companies race to build machines that can perform physical tasks in the real world, they face a critical bottleneck: a shortage of high-quality, real-world training data showing humans doing ordinary work. Human Archive’s central thesis is that the workers staffing India’s rapidly expanding gig economy represent an untapped and scalable source of exactly that data.<\/p>\n

Why Indian gig workers are becoming a prime source for robot training data<\/h2>\n

India’s online food delivery market has grown significantly in recent years, with major players like Zomato and Swiggy going public and a proliferation of cloud kitchens. Simultaneously, startups focused on home services \u2014 on-demand household staffing platforms such as Urban Company, Snabbit, and Pronto \u2014 have gained substantial traction. Human Archive is tapping directly into this ecosystem, partnering with companies in the home services, hostel, and restaurant sectors to deploy more than 1,000 active headsets across multiple locations, capturing egocentric video data of workers performing routine tasks.<\/p>\n

The company’s founding team brings research backgrounds spanning robotics, hardware, and tactile data, which informs their belief that video data alone is insufficient for training robust physical AI systems. To differentiate itself from other data collection efforts, Human Archive is developing and deploying a suite of additional sensing devices, including tactile gloves, full-body motion capture suits, and wrist cameras. These instruments capture motion data and tactile force synchronously aligned with RGB-D imagery \u2014 color video paired in real time with depth information \u2014 creating multi-modal datasets that the startup sells to AI labs.<\/p>\n

Building custom hardware for multi-modal data collection<\/h2>\n

The company’s hardware journey reflects its iterative approach to solving the data bottleneck. Initially, Human Archive relied on makeshift setups and off-the-shelf rigs, including iPhones, to capture footage. Over time, the team built custom rigs and caps, and now deploys more than seven different hardware products that can be used interchangeably across different sensing modalities. In total, the company has more than 50 different devices deployed to collect diverse data points.<\/p>\n

“To capture data, we started with iPhones, then we built our own custom rigs and caps. Now we have more than seven different hardware products that we use interchangeably across different modalities,” CEO Raj Patel said in a call. “After data collection from different devices, we worked on synchronizing data from all these different sources.”<\/p>\n

This multi-sensor approach has attracted significant interest from the research community. Zach DeWitt, a partner at Wing VC, emphasized the uniqueness of Human Archive’s capability. “No one else in the world has been able to synchronize and collect headset RGB-D, force feedback, full-body motion capture, and synchronized chest and wrist camera data at scale,” DeWitt told TechCrunch. “They’ve been doing internal model training on this data, and every major lab and university is interested in running experiments on it due to the novelty of the sensors and the scale of the new dataset they are releasing soon.”<\/p>\n

Human Archive is also developing ways to fine-tune AI models using its own data and testing them on robots to evaluate task effectiveness, allowing the startup to demonstrate data quality to potential customers and post-train internal models.<\/p>\n

Rejection by major home services platforms and a very public feud<\/h2>\n

Despite its traction, Human Archive has faced significant resistance from some of India’s largest home services companies. The startup acknowledged that it was rejected by many potential partners, including Pronto and Urban Company. That rejection became public fodder when Indian outlet Entrackr reported that Pronto was actively seeking partnerships to collect worker data for robotics training and that Snabbit had held early discussions with Human Archive before the project fell apart.<\/p>\n

Urban Company CEO Abhiraj Singh Bhal responded on X, stating unequivocally that his company would not engage in such arrangements. That post prompted Patel to fire back that Urban Company would soon be forced to reconsider or risk losing relevance to customer churn. Co-founder Rushil Agarwal was blunter still, posting that Pronto founder Anjali Sardana had laughed at him and called him “stupid” when he raised the idea of a data partnership. Pronto acknowledged the conversations but said it chose not to move forward.<\/p>\n

The public exchange highlights the tensions inherent in a business model that turns workers’ physical labor into training data for machines that could eventually automate the same tasks. It also underscores the competitive dynamics in India’s home services sector, where data ownership and worker privacy are becoming flashpoints.<\/p>\n

The business model: discounted services for data, and dollar-an-hour wages<\/h2>\n

Despite rejection by larger players, Human Archive has teamed up with smaller startups to offer discounted services to end customers. When a worker arrives at a home, consumers are given a choice through the app: pay a reduced price in exchange for consenting to data collection, or pay the full price for an unrecorded visit. Patel noted that customers have been happy to opt for the former, as disputes about service quality are common and video recordings can help resolve them.<\/p>\n

The company pays workers a base rate of $1 per hour for participating in egocentric data collection. By comparison, a report from the Economic Times suggests that other companies pay between 250 and 400 rupees per hour, roughly $2.63 to $4.20. Patel acknowledged that competitors pay more but argued that Human Archive’s on-the-ground presence in India allows it to keep compensation lower. Wing VC’s DeWitt framed the wage structure as a feature, not a bug, stating that the company’s “network provides immediate, flexible earning opportunities globally, lowering the barrier to participating in the AI economy. We see this as a critical bridge that funds immediate livelihoods while building the infrastructure for a safer, more productive future.”<\/p>\n

Privacy concerns and regulatory scrutiny<\/h2>\n

The data collection model raises significant privacy questions. It remains unclear what information Human Archive provides workers about how their footage is used. The company maintains that its commercial contracts comply with India’s Digital Personal Data Protection Act, displaying a privacy policy notice along with consent information detailing the purpose of data collection and how it is processed. The company says all data is anonymized and that faces are blurred from recordings.<\/p>\n

Nevertheless, the broader industry is drawing regulatory attention. Moneycontrol reported that India’s Ministry of Electronics and Information Technology is looking into the consent mechanisms and data collection practices of startups collecting egocentric data through home service workers. That scrutiny reflects growing global concern about the use of workers as data sources for AI training, particularly when compensation is low and the long-term implications for job displacement are unresolved.<\/p>\n

Expansion beyond India and the race for physical AI data<\/h2>\n

While Human Archive currently collects the majority of its data in India, the company has begun expanding into Southeast Asia and the United States. It is building a platform that would allow anyone to participate in data collection and earn money, and it also wants to offer customers in the U.S. services like cleaning or cooking in exchange for data collection by participating workers \u2014 though these programs remain in an early pilot stage.<\/p>\n

Human Archive is entering a crowded and well-funded field. Multiple startups are racing to build physical AI systems, and doing so requires massive amounts of training data showing humans at work. Whether Human Archive’s approach can scale will hinge on the partnerships it can strike and the uniqueness and volume of the data it can collect to satisfy the appetite of physical AI labs. The company’s multi-sensor approach and growing hardware portfolio give it a differentiated position, but the rejection by major platforms, regulatory headwinds, and ethical questions around worker compensation and privacy present significant challenges.<\/p>\n

The bet that India’s gig workers can serve as the eyes and hands for the next generation of robots is both ambitious and fraught. Human Archive has placed itself at the center of that bet, and its $8.2 million raise suggests that investors believe the payoff could be substantial \u2014 even if the path forward is anything but simple.<\/p>\n","protected":false},"excerpt":{"rendered":"

In a funding round that underscores the escalating demand for physical AI training data, Silicon Valley-based Human Archive has announced it raised $8.2 million from a consortium that includes Wing Venture Capital, NVP Capital, Y Combinator, and angel investors affiliated with OpenAI, Nvidia, Google, Mercor, AfterQuery, BAIR, SAIL, and Meta. The startup is building a […]<\/p>\n","protected":false},"author":6,"featured_media":84949,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/53722.png","fifu_image_alt":"Human Archive Raises $8.2 Million to Train Robots Using Indian Gig Workers","footnotes":""},"categories":[31],"tags":[],"class_list":["post-53722","post","type-post","status-publish","format-standard","has-post-thumbnail","category-technology"],"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/53722.png","fifu_image_alt":"Human Archive Raises $8.2 Million to Train Robots Using Indian Gig Workers","_links":{"self":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/53722","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/comments?post=53722"}],"version-history":[{"count":0,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/53722\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media\/84949"}],"wp:attachment":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media?parent=53722"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/categories?post=53722"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/tags?post=53722"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}