Meet Ganesh Venkataramanan, the ex-Tesla Dojo chief secretly building DensityAI with 20 former teammates
Around 20 ex-Tesla Dojo engineers have joined DensityAI
Startup aims to simplify costly AI infrastructure for carmakers
A stealthy new player has just pulled into the fast lane of the automotive AI race, and it's being driven by a man who knows a thing or two about horsepower in the digital age. Ganesh Venkataramanan, the former mastermind behind Tesla's Dojo supercomputer project, has resurfaced at the wheel of DensityAI, a startup aiming to supercharge how automakers train and deploy self-driving technology. And he hasn't come alone, around 20 of his old Tesla colleagues, many senior members of the Dojo engineering squad, are riding shotgun.
The mission? Build a full-stack AI platform, from chips to software, designed specifically for the unique demands of the automotive world. Unlike generic AI systems, DensityAI wants to create hardware and tools optimised for the immense data-crunching required by autonomous vehicles, covering everything from sensor fusion and simulation to edge computing. From Tesla's Dojo to DensityAI
Venkataramanan left Tesla in late 2023 after a seven-year stint that saw him lead Dojo, the custom-built supercomputer used to train Tesla's neural networks on a flood of driving footage. Before that, he honed his chip-design credentials at AMD. Now, he's channelling that expertise into a new platform that could remove one of the biggest headaches for carmakers: building and maintaining costly AI infrastructure in-house.
Self-driving technology demands colossal computing muscle, and the data centres needed to power it are often as complex to manage as the cars themselves. DensityAI is positioning itself as the plug-and-play solution â€' delivering high-performance, high-density computing tailored for the realities of the road.
Nvidia might dominate the automotive AI chip market today. Still, several reports claim DensityAI could grab a chunk of the action by zooming in on automotive-specific needs, instead of the one-size-fits-all approach favoured by bigger players. Early whispers suggest the company is already in talks with manufacturers and eyeing a major funding round worth hundreds of millions as it emerges from stealth mode.
While the startup's primary focus is on cars, its ambitions stretch further. Robotics and other industries reliant on real-time AI processing are also in its sights, with the broader aim of creating data centre infrastructure that can scale efficiently across sectors.
Venkataramanan's departure is just the latest in a line of high-profile exits from Tesla's AI ranks. In 2023, Andrej Karpathy, the company's former AI director, headed back to OpenAI. Others have left to start ventures of their own or voice concerns about the complexity and safety hurdles of delivering true autonomy.
Tesla itself isn't slowing down, it's reportedly teaming up with Samsung on its next-gen AI chips, but the flow of talent into new, specialised startups like DensityAI highlights shifting dynamics in the sector.
Industry experts caution that breaking into this market isn't just about clever engineering. Developing reliable, scalable AI systems for vehicles is a long game, and the road is lined with regulatory speed bumps. Automakers face intense scrutiny when it comes to safety, testing, and deployment, making trust as important as raw performance.
Even so, DensityAI's founders have a rare advantage: they've already built one of the most sophisticated AI training platforms on the planet. If they can apply that same focus to solving the bottlenecks facing carmakers, they could help speed up the rollout of safer, smarter autonomous systems.
Observers on X have dubbed DensityAI a 'full-stack AI data centre company', meaning it plans to compete on both the hardware and the software fronts. That holistic approach could prove to be a game-changer, giving automakers a one-stop shop for training, deploying, and maintaining their AI.
DensityAI's first products, expected to launch in the coming months, will aim to help carmakers train and deploy AI faster, more intelligently, and more cost-effectively. If Venkataramanan and his crew pull it off, they won't just be tweaking how cars think, they could redefine how the entire industry processes data.
After all, this is a team that's already built a record-breaking AI system once. Now, they're betting they can do it again, and this time, the whole automotive world might be along for the ride.
A stealthy new player has just pulled into the fast lane of the automotive AI race, and it's being driven by a man who knows a thing or two about horsepower in the digital age. Ganesh Venkataramanan, the former mastermind behind Tesla's Dojo supercomputer project, has resurfaced at the wheel of DensityAI, a startup aiming to supercharge how automakers train and deploy self-driving technology. And he hasn't come alone, around 20 of his old Tesla colleagues, many senior members of the Dojo engineering squad, are riding shotgun.
The mission? Build a full-stack AI platform, from chips to software, designed specifically for the unique demands of the automotive world. Unlike generic AI systems, DensityAI wants to create hardware and tools optimised for the immense data-crunching required by autonomous vehicles, covering everything from sensor fusion and simulation to edge computing. From Tesla's Dojo to DensityAI
Venkataramanan left Tesla in late 2023 after a seven-year stint that saw him lead Dojo, the custom-built supercomputer used to train Tesla's neural networks on a flood of driving footage. Before that, he honed his chip-design credentials at AMD. Now, he's channelling that expertise into a new platform that could remove one of the biggest headaches for carmakers: building and maintaining costly AI infrastructure in-house.
Self-driving technology demands colossal computing muscle, and the data centres needed to power it are often as complex to manage as the cars themselves. DensityAI is positioning itself as the plug-and-play solution â€' delivering high-performance, high-density computing tailored for the realities of the road.
Nvidia might dominate the automotive AI chip market today. Still, several reports claim DensityAI could grab a chunk of the action by zooming in on automotive-specific needs, instead of the one-size-fits-all approach favoured by bigger players. Early whispers suggest the company is already in talks with manufacturers and eyeing a major funding round worth hundreds of millions as it emerges from stealth mode.
While the startup's primary focus is on cars, its ambitions stretch further. Robotics and other industries reliant on real-time AI processing are also in its sights, with the broader aim of creating data centre infrastructure that can scale efficiently across sectors.
Venkataramanan's departure is just the latest in a line of high-profile exits from Tesla's AI ranks. In 2023, Andrej Karpathy, the company's former AI director, headed back to OpenAI. Others have left to start ventures of their own or voice concerns about the complexity and safety hurdles of delivering true autonomy.
Tesla itself isn't slowing down, it's reportedly teaming up with Samsung on its next-gen AI chips, but the flow of talent into new, specialised startups like DensityAI highlights shifting dynamics in the sector.
Industry experts caution that breaking into this market isn't just about clever engineering. Developing reliable, scalable AI systems for vehicles is a long game, and the road is lined with regulatory speed bumps. Automakers face intense scrutiny when it comes to safety, testing, and deployment, making trust as important as raw performance.
Even so, DensityAI's founders have a rare advantage: they've already built one of the most sophisticated AI training platforms on the planet. If they can apply that same focus to solving the bottlenecks facing carmakers, they could help speed up the rollout of safer, smarter autonomous systems.
Observers on X have dubbed DensityAI a 'full-stack AI data centre company', meaning it plans to compete on both the hardware and the software fronts. That holistic approach could prove to be a game-changer, giving automakers a one-stop shop for training, deploying, and maintaining their AI.
DensityAI's first products, expected to launch in the coming months, will aim to help carmakers train and deploy AI faster, more intelligently, and more cost-effectively. If Venkataramanan and his crew pull it off, they won't just be tweaking how cars think, they could redefine how the entire industry processes data.
After all, this is a team that's already built a record-breaking AI system once. Now, they're betting they can do it again, and this time, the whole automotive world might be along for the ride. Join our WhatsApp Channel
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Time of India
4 hours ago
- Time of India
Role of Brand Advocacy in B2B Influencer Marketing
By Ajit Narayan Influencer Marketing - while the words are the same, the usage and application of it are completely different in B2B and B2C. In B2C, one is after brand alignment and emotional appeal, which drives purchasing decisions whereas in B2B, the priority is in category expertise, reliability and category following. And the Influencers are also different. B2B influencers are typically industry experts, leaders themselves and customers who have a voice and bring in opinions in the industry. They play a big role in influencing decisions by companies, be it in hiring or taking on new clients and partnerships. Typically, in B2C purchase or even evaluation, decisions are often impulsive. In B2B, the buying process is complex and involves multiple levels and departments before making a purchase decision, and therefore the expertise of influencers and their reach plays a vital role. Combined with Brand Advocacy , the reach of B2B influencers makes a big difference in reach and engagement as the objectives are aligned: Building Trust and Credibility B2B buyers rely on trusted sources to navigate complex purchasing decisions. Advocates, such as industry professionals or existing customers, lend authenticity that traditional advertising lacks. Employees, Customers or Partners share genuine experiences, case studies, or insights about a brand's product/service, reinforcing its value. As per a Demand Gen Report 2023, 92% of B2B buyers trust recommendations from peers over branded content. Amplifying Reach and Engagement Advocates often have their own established personal networks, giving brands access to niche, high-value audiences. Content shared by advocates generates 2-3x higher engagement rates than brand-owned channels, according to a 2024 LinkedIn study. Humanising the Brand B2B brands can seem impersonal. Brand Advocates humanize them by sharing relatable stories or practical use cases. A customer post about how a product solved a specific pain point, or an industry influencer could discuss trends aligning with the brand's mission is an authentic voice on the use cases. Personalized advocacy content drives 20-30% higher conversion rates in B2B campaigns (per 2023 Influitive data). Drives Thought Leadership B2B buyers seek partners who demonstrate expertise. Advocates position brands as industry leaders by associating them with credible voices. Influencers co-create content like reports, podcasts, or panel discussions, showcasing the brand's innovation or problem-solving capabilities. 74% of B2B buyers consume thought leadership content before making purchase decisions (Edelman, 2024). Facilitating Long-Term Relationships B2B sales cycles are long, often spanning months. Advocates nurture prospects by maintaining consistent, credible engagement. Advocates participate in ongoing campaigns, such as speaking at webinars, contributing to case studies, or engaging in community discussions on social media platforms. Companies with strong advocacy programs see 25% shorter sales cycles (per 2023 Forrester study). In B2B Marketing , both brand advocacy and influencer marketing tango together. And many a times could be one and the same as well. As one could leverage their own key stake holders in the marketing mix to develop and share content favoring the brand. Brand Advocacy involves creating strategic program involving key stakeholders in the mix and a good tactical mix would be to include Influencers as interventions to boost the existing program. In a nutshell it is only beneficial to think of both in a complementary sense rather than different practices in the marketing mix for B2B as they are very similar. Only that the Influencer Marketing is interventional. Whereas advocacy is consistent through the year. (The author is CMO, Socxo. Opinions are personal.)
Time of India
5 hours ago
- Time of India
LinkedIn launches new casual game for its users; company executive says: ‘We don't want to have…'
LinkedIn has launched a new, casual game, Mini Sudoku, for its 1.2 billion users. The Microsoft-owned professional networking site's latest addition is a scaled-down version of the classic puzzle, designed to be completed in just two or three minutes. This is the sixth game to be introduced on the platform. The new Mini Sudoku aims to spark friendly competition among colleagues and friends, with puzzles getting progressively more difficult throughout the week. What the company said about the new Mini Sudoku game In an interview with CNBC, Lakshman Somasundaram , a senior director of product at the company said: 'We don't want to have a puzzle on LinkedIn that takes 20 minutes to solve, right? We're not games for games' sake.' LinkedIn's creation of the game stemmed from a meeting with Japanese publisher Nikoli , known for popularising Sudoku. Last year, Somasundaram and a group of LinkedIn associate product managers visited Nikoli's Tokyo headquarters, where they discussed puzzles with the publisher's employees through a translator. This meeting led to weeks of collaboration between LinkedIn, Nikoli, and Thomas Snyder , a three-time World Sudoku Championship winner who has been advising LinkedIn on its gaming strategy. The team aimed to make Sudoku more approachable, experimenting with several prototypes before deciding on a board featuring six rows and six columns. 'It's very easy to just make a Sudoku grid. It's very hard to make art in the form of Sudoku. And that's what both Nikoli and we do. I think it's got the potential to be the largest of the games, just because it's going to have a lot of brand awareness from moment one,' Snyder noted. Snyder, who is the founder and CEO of Grandmaster Puzzles, a publisher of Sudoku books, holds a PhD in chemistry and is known as Dr Sudoku. He has worked on the hint feature for LinkedIn's Mini Sudoku and created some of the puzzles. Each day's puzzle will be accompanied by a video of Snyder demonstrating his solving process. However, this is not the first game LinkedIn has introduced. The platform added games last year to bring a sense of fun and offer users fresh ways to engage with each other. According to a company spokesperson, millions of people play LinkedIn's games daily, with peak activity at 7 AM ET (4.30 PM IST) and Gen Z make up the largest share of players. Among the ones who play on a given day, 86% return the next day, and 82% are still playing a week later. UBON SP-85: Portable Party Speaker On A Budget AI Masterclass for Students. Upskill Young Ones Today!– Join Now
Time of India
7 hours ago
- Time of India
Rs 80,000 EMI or Rs 35,000 rent? One Reddit post has everyone rethinking their life and financial choices
Internet debates What does financial analyst say? It's the kind of question that can split a dinner table right down the middle — is it smarter to buy a house with a hefty loan or rent and invest the difference? A viral Reddit post has sparked a heated discussion, with people weighing in with personal stories, hard numbers, and strong opinions on what makes the most financial sense in today's debate kicked off when a Reddit user urged people not to take a loan for buying a flat, suggesting they rent instead. According to him, any semi-decent flat in a tier 1 city costs at least Rs 1.5 crore. Even with a 20% down payment of Rs 30 lakh, the loan amount would still be Rs 1.2 crore. At current home loan rates of about 7.5% per annum, he calculated an EMI of Rs 80,000 per month — stretching all the way to retirement. He argued that the total interest paid would equal the value of the flat, meaning you're effectively a tenant to the bank until the loan is cleared. 'You're not the owner. You'll be kicked out if you lose your job and miss your EMI payments,' he wrote, adding that flats depreciate over time and aren't the great inheritance many believe them to be. In contrast, the same property could be rented for Rs 30,000–Rs 35,000 a month, leaving room to save and invest the post set off a flurry of responses. One user agreed that owning a flat in the current unstable environment sounded risky, noting that two decades ago it would have been a no-brainer. Another pushed back, saying a Rs 1.5 crore flat fetching only Rs 35,000 in rent was unrealistic in most job-rich cities, and that home ownership is about more than numbers. They argued that many finance influencers advising against buying come from secure family backgrounds — for someone with little to no inheritance, owning a home could be a top highlighted the practical benefits of ownership. One pointed out that as long as you pay your EMIs, a bank can't evict you — unlike a landlord, who can ask you to vacate with notice, even on a whim. They noted that homeowners can legally rent out their property, make structural changes, and participate in housing society decisions — rights that most tenants lack. They also mentioned that EMIs tend to be influenced by repo rates, while rents can be some users maintained that buying only makes sense if the flat is well within your means. 'Never buy a flat which takes almost your full or even 75% of your take-home salary,' one warned. Their advice: buy a cheaper flat in a lower locality, or even in a tier 3 city, just to ensure you always have a home to fall back Hardik Joshi, a Hyderabad-based financial analyst, had shared in a LinkedIn post that buying suits for those with stable jobs and deep roots, while renting benefits those seeking mobility, financial growth, and global. He weighed in on the rent vs buy debate, noting that purchasing a home in metros like Mumbai, Delhi NCR, Bengaluru, or Hyderabad demands a heavy financial commitment. A 2BHK in a prime area can cost Rs 1.5 crore, with a Rs 30 lakh down payment and a Rs 1.2 crore loan at 9% interest, leading to repayments exceeding Rs 2.5 crore over 20 years. Beyond the cost, buying limits mobility, making it harder to pursue better opportunities elsewhere. Homeowners also bear ongoing expenses like maintenance, property taxes, and face market risks that could affect resale value, making ownership as much a lifestyle choice as a financial one.
