{"id":3488987,"date":"2026-03-31T15:00:00","date_gmt":"2026-03-31T15:00:00","guid":{"rendered":"https:\/\/techingeek.com\/index.php\/2026\/03\/31\/nomadic-secures-8-4-million-to-manage-the-data-streaming-from-self-driving-cars\/"},"modified":"2026-03-31T15:00:00","modified_gmt":"2026-03-31T15:00:00","slug":"nomadic-secures-8-4-million-to-manage-the-data-streaming-from-self-driving-cars","status":"publish","type":"post","link":"https:\/\/techingeek.com\/index.php\/2026\/03\/31\/nomadic-secures-8-4-million-to-manage-the-data-streaming-from-self-driving-cars\/","title":{"rendered":"Nomadic secures $8.4 million to manage the data streaming from self-driving cars."},"content":{"rendered":"
<\/div>\n

To create the autonomous machines of tomorrow, sometimes a model is required for your model.\u00a0<\/p>\n

Businesses that are working on self-driving vehicles, robots interacting with the physical world, or autonomous construction machinery gather thousands, if not millions, of hours of video footage for analysis and training.\u00a0<\/p>\n

Sorting and archiving that video is currently a task for humans, who must view all of it. Even when fast-forwarding, that process doesn’t scale. NomadicML, a company established by CEO Mustafa Bal and CTO Varun Krishnan, aims to address issues for clients who have 95% of their fleet data stored away.<\/p>\n

The difficulty increases when searching for edge cases \u2014 the most crucial data showcases events that seldom happen and can confuse inexperienced physical AI models.<\/p>\n

Nomadic is aiming to tackle that issue with a platform that converts footage into a structured, searchable dataset through a suite of vision language models. This, in turn, enables improved fleet monitoring and the crafting of distinctive datasets for reinforcement learning and quicker iteration.<\/p>\n

The company revealed an $8.4 million seed funding round on Tuesday at a post-money valuation of $50 million. This round was led by TQ Ventures, with contributions from Pear VC and Jeff Dean, and will facilitate the onboarding of more clients and the ongoing enhancement of its platform. Nomadic also secured first place in Nvidia GTC\u2019s pitch contest last month.\u00a0<\/p>\n

The two founders, who met as undergraduates in computer science at Harvard, “kept encountering the same technical obstacles repeatedly at our jobs” at firms such as Lyft and Snowflake, Bal shared with TechCrunch.\u00a0<\/p>\n

\n
\n

Techcrunch event<\/p>\n

\n

\n\t\t\t\t\t\t\t\t\tSan Francisco, CA<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\t|<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\tOctober 13-15, 2026<\/span>\n\t\t\t\t\t\t\t<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n

\u201cWe offer individuals insights into their own footage, whatever drives their own AVs [and] robots,\u201d he stated. \u201cThat is what propels these developers of autonomous systems forward, not arbitrary data.\u201d<\/p>\n

Consider, for instance, trying to enhance an AV\u2019s comprehension that it can proceed through a red light if a police officer commands it to do so, or identifying every instance when vehicles pass under a particular type of bridge. Nomadic\u2019s platform enables these occurrences to be recognized for both compliance reasons and to be directly integrated into training pipelines.\u00a0<\/p>\n

Clients such as Zoox, Mitsubishi Electric, Natix Network, and Zendar are already leveraging the platform to create intelligent machines. Antonio Puglielli, the VP of Engineering at Zendar, remarked that Nomadic\u2019s tool enabled the company to accelerate its work much more swiftly than the option of outsourcing, and that its domain expertise distinguished it from other competitors.<\/p>\n

This type of model-based, auto-annotation technology is emerging as a vital workflow for physical AI. Established data labeling companies like Scale, Kognic, and Encord are creating AI tools to conduct this work, while Nvidia has launched a series of open-source models, Alpamayo, that can be tailored to address the issue.<\/p>\n

Varun contends that his company\u2019s tool goes beyond mere labeling; it functions as an \u201cagentic reasoning system: you specify what it needs, and it determines how to find it,\u201d utilizing multiple models to comprehend actions occurring and contextualize them. Nomadic\u2019s investors anticipate that the startup\u2019s focus on this specific infrastructure will prevail.\u00a0<\/p>\n

\u201cIt\u2019s akin to why Salesforce doesn\u2019t create its own cloud and Netflix doesn\u2019t construct its own [content distribution facilities],\u201d Schuster Tanger, a partner at TQ Ventures who spearheaded the round, told TechCrunch. \u201cThe moment an autonomous vehicle company attempts to internally develop Nomadic, they divert from what enables their success, which is the robot itself.\u201d<\/p>\n

Tanger commends Nomadic\u2019s talent, highlighting that Krishnan is an international chess master ranked as the world\u2019s 1,549th-best player. Krishnan, in turn, boasts that all of the company\u2019s roughly dozen engineers have published scientific papers.<\/p>\n

Now, they are diligently working on specific tools, like one that comprehends the physics of lane changes from camera footage, or another that derives more precise placements for a robot\u2019s grippers in a video. The forthcoming challenge, from the perspective of Nomadic and its clients, is to develop similar tools for non-visual data like lidar sensor readings or to unify sensor data across various modes.\u00a0<\/p>\n

\u201cHandling terabytes of video, processing that against hundreds of 100 billion-plus parameter models, and then extracting their accurate insights is exceedingly challenging,\u201d Bal stated.\u00a0<\/p>\n","protected":false},"excerpt":{"rendered":"

<\/div>\n

To create the autonomous machines of tomorrow, sometimes a model is required for your model.\u00a0<\/p>\n

Businesses that are working on self-driving vehicles, robots interacting with the physical world, or autonomous construction machinery gather thousands, if not millions, of hours of video footage for analysis and training.\u00a0<\/p>\n

Sorting and archiving that video is currently a task for humans, who must view all of it. Even when fast-forwarding, that process doesn’t scale. NomadicML, a company established by CEO Mustafa Bal and CTO Varun Krishnan, aims to address issues for clients who have 95% of their fleet data stored away.<\/p>\n

The difficulty increases when searching for edge cases \u2014 the most crucial data showcases events that seldom happen and can confuse inexperienced physical AI models.<\/p>\n

Nomadic is aiming to tackle that issue with a platform that converts footage into a structured, searchable dataset through a suite of vision language models. This, in turn, enables improved fleet monitoring and the crafting of distinctive datasets for reinforcement learning and quicker iteration.<\/p>\n

The company revealed an $8.4 million seed funding round on Tuesday at a post-money valuation of $50 million. This round was led by TQ Ventures, with contributions from Pear VC and Jeff Dean, and will facilitate the onboarding of more clients and the ongoing enhancement of its platform. Nomadic also secured first place in Nvidia GTC\u2019s pitch contest last month.\u00a0<\/p>\n

The two founders, who met as undergraduates in computer science at Harvard, “kept encountering the same technical obstacles repeatedly at our jobs” at firms such as Lyft and Snowflake, Bal shared with TechCrunch.\u00a0<\/p>\n

\n
\n

Techcrunch event<\/p>\n

\n

\n\t\t\t\t\t\t\t\t\tSan Francisco, CA<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\t|<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\tOctober 13-15, 2026<\/span>\n\t\t\t\t\t\t\t<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n

\u201cWe offer individuals insights into their own footage, whatever drives their own AVs [and] robots,\u201d he stated. \u201cThat is what propels these developers of autonomous systems forward, not arbitrary data.\u201d<\/p>\n

Consider, for instance, trying to enhance an AV\u2019s comprehension that it can proceed through a red light if a police officer commands it to do so, or identifying every instance when vehicles pass under a particular type of bridge. Nomadic\u2019s platform enables these occurrences to be recognized for both compliance reasons and to be directly integrated into training pipelines.\u00a0<\/p>\n

Clients such as Zoox, Mitsubishi Electric, Natix Network, and Zendar are already leveraging the platform to create intelligent machines. Antonio Puglielli, the VP of Engineering at Zendar, remarked that Nomadic\u2019s tool enabled the company to accelerate its work much more swiftly than the option of outsourcing, and that its domain expertise distinguished it from other competitors.<\/p>\n

This type of model-based, auto-annotation technology is emerging as a vital workflow for physical AI. Established data labeling companies like Scale, Kognic, and Encord are creating AI tools to conduct this work, while Nvidia has launched a series of open-source models, Alpamayo, that can be tailored to address the issue.<\/p>\n

Varun contends that his company\u2019s tool goes beyond mere labeling; it functions as an \u201cagentic reasoning system: you specify what it needs, and it determines how to find it,\u201d utilizing multiple models to comprehend actions occurring and contextualize them. Nomadic\u2019s investors anticipate that the startup\u2019s focus on this specific infrastructure will prevail.\u00a0<\/p>\n

\u201cIt\u2019s akin to why Salesforce doesn\u2019t create its own cloud and Netflix doesn\u2019t construct its own [content distribution facilities],\u201d Schuster Tanger, a partner at TQ Ventures who spearheaded the round, told TechCrunch. \u201cThe moment an autonomous vehicle company attempts to internally develop Nomadic, they divert from what enables their success, which is the robot itself.\u201d<\/p>\n

Tanger commends Nomadic\u2019s talent, highlighting that Krishnan is an international chess master ranked as the world\u2019s 1,549th-best player. Krishnan, in turn, boasts that all of the company\u2019s roughly dozen engineers have published scientific papers.<\/p>\n

Now, they are diligently working on specific tools, like one that comprehends the physics of lane changes from camera footage, or another that derives more precise placements for a robot\u2019s grippers in a video. The forthcoming challenge, from the perspective of Nomadic and its clients, is to develop similar tools for non-visual data like lidar sensor readings or to unify sensor data across various modes.\u00a0<\/p>\n

\u201cHandling terabytes of video, processing that against hundreds of 100 billion-plus parameter models, and then extracting their accurate insights is exceedingly challenging,\u201d Bal stated.\u00a0<\/p>\n","protected":false},"author":2,"featured_media":3488988,"comment_status":"open","ping_status":"closed","sticky":false,"template":"Default","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-3488987","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/posts\/3488987"}],"collection":[{"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/comments?post=3488987"}],"version-history":[{"count":0,"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/posts\/3488987\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/media\/3488988"}],"wp:attachment":[{"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/media?parent=3488987"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/categories?post=3488987"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/techingeek.com\/index.php\/wp-json\/wp\/v2\/tags?post=3488987"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}