EnCharge AI is a leader in advanced hardware and software systems for AI computing. EnCharge’s robust and scalable next-generation in-memory computing technology provides orders-of-magnitude higher compute efficiency and density compared to today’s best-in-class solutions, at a fraction of the cost. The high-performance architecture is coupled with seamless software and will enable the immense potential of AI to be accessible in power -, energy-, and space-constrained applications. EnCharge AI launched in 2022 and is led by veteran technologists with backgrounds in semiconductor design and AI systems.
Thank you so much, Princeton ECE, for highlighting EnCharge AI for this year's Celebrate Princeton Innovation. We are grateful for your continued support. The work we're doing to radically transform AI energy efficiency wouldn't be possible without you.
“Princeton has strengths across the disciplines in the humanities, social sciences, engineering and natural sciences like exists nowhere else on the planet,” Naveen Verma said at this year's Celebrate Princeton Innovation. “If we think about the technologies that we’re building, these technologies we see today are going to have tremendous impact.”
The evening’s featured company was EnCharge AI, a startup based on Princeton research that develops computer chips that can store data and run artificial intelligence computations more affordably and more quickly while consuming less energy.
Verma, company founder and CEO and the Ralph H. and Freda I. Augustine Professor of Electrical and Computer Engineering, said it’s an exciting time to be a faculty member and an entrepreneur.
Read the full story for more details, including highlights of faculty working on innovations in computer vision and critical minerals 👉 https://lnkd.in/e95dKJKV
✍️ Dan Day
📷 Sameer A. Khan/Fotobuddy
The U.S. is facing a major national security threat: losing ground on the AI-defined battlefield. Not because we lack cutting-edge models, but because we lack the ultra-efficient chips needed to run them at the tactical edge.
A recent Center for Strategic and International Studies (CSIS) report highlights the gap. The U.S. has invested heavily in large cloud-based models that run in data centers. They’re powerful, but unusable in environments where networks are insecure, latency is life-threatening, and energy is limited. What’s missing is the right kind of AI: edge AI compute that runs inference directly on the devices where data is generated.
That shift is finally possible. New chip architectures, including analog in-memory compute, which we’re advancing at EnCharge AI, are delivering significant efficiency gains that bring advanced inference directly onto devices for the first time. This is what will enable the next generation of battlefield capabilities: autonomous drone swarms, ground robots that respond instantly to threats, and systems that remain reliable with zero dependence on data centers.
The technology exists today. What’s holding us back is not development, but deployment. The U.S. must move immediately to integrate edge AI into defense systems, because the advantage will go to whoever fields it first.
#NationalSecurity#Defense#EdgeAI#EfficientAIhttps://lnkd.in/eBtgg_DS
The AI infrastructure crisis is unfolding faster than anyone expected. New data from Bain & Company shows data centers will consume 9% of US electricity by 2030; more than double today’s share. This is not just a capacity issue. It is a fundamental architectural problem at the heart of AI compute.
The culprit is clear. Today’s GPU-centric designs waste 80-95% of inference energy simply moving data between memory and compute. This “commute” is now AI’s biggest bottleneck, one that building more multi-gigawatt data centers cannot fix.
At EnCharge AI, we're addressing this bottleneck at its source. Our analog in-memory computing technology processes data where it lives, targeting up to 20x efficiency improvements, among the highest in the industry. By attacking the root cause of AI's energy crisis, we're developing the foundation for next-generation AI applications without unsustainable power demands.
While the industry rushes to restart nuclear plants and pour billions into new infrastructure, our focus is on eliminating the von Neumann bottleneck that makes current inference architectures fundamentally inefficient. EnCharge AI’s EN100 chip, launching in early 2026, marks a critical step toward this future.
The next era of AI will not be defined by power alone. It will be defined by how intelligently we use it.
https://lnkd.in/d46hT2qr#AIInfrastructure#Sustainability#InMemoryComputing#DataCenters#EnergyEfficiency
We’re excited to share that EnCharge AI co-founder & CEO Naveen Verma will deliver a keynote as featured innovator at Celebrate Princeton University Innovation 2025!
The annual event highlights Princeton faculty whose discoveries are shaping everyday innovations that improve lives, benefit the planet, and fuel the economy.
Join us tomorrow, Oct 22 (5:00-7:30pm ET at Frick Chemistry Lab) as Naveen explores how EnCharge's breakthrough architecture is redefining how and where advanced AI can be deployed.
Register: https://lnkd.in/evGxKvh8
“If you think a breakthrough alone will carry the company forward, you’re mistaken.” I was struck by how candid and insightful that statement was when I talked with Naveen Verma, CEO of EnCharge AI, a couple weeks ago. It instantly gets to the heart of what it takes to grow a company centered on innovation, and confirms long-held beliefs I have about startups born in academia.
The breakthrough he referenced is EnCharge’s EN100, recently released following their $100M Series B raise. Unlike companies focused on digital architectures, EnCharge is leveraging analog for inference.
Once confined to research labs, analog is gaining fresh attention, and EnCharge’s news signaled their approach is ready for primetime. But turning differentiated technology into commercial success (not just research outcomes) takes massive effort, especially in attracting the people to build it.
“Everything that we do has to be built and designed in a way that’s optimized around a very new way of fundamentally doing computation,” Naveen explained. “That's a change that comes from within, and it pushes you out of your comfort zone.”
This highlights an ethos I’ve long believed and have trumpeted from the mountaintops: building an organization fit for the long run requires “repelling the many and compelling the few.” Leaders MUST be radically transparent about the challenges their teams will face, and unwavering on who aligns with the mission and who doesn't. Naveen clearly knows this.
Speaking of breakthroughs, this well-respected professor is one of the few in memory who seems to have found the key to being both an academic and commercial leader.
I call it a breakthrough because, among a graveyard of startups founded by researchers who remained professors while becoming founders, most have failed. Those leaders couldn’t strike the balance between academic preoccupations and delivering products to market. Naveen understands how to do this, and he explained the lessons he's learned in another part of our conversation.
With so much talk about ‘work-life balance', if you’re truly doing your life’s work, the line disappears. That inner drive to build something world-changing shouldn’t compete with other distractions or demands, and Naveen does it all while making time for the most important work of all: his family.
The way he lives clearly shapes how he leads EnCharge, which should be a beacon for anyone drawn to startups with a culture of obsessive execution, but fear the 'romanticized' 18-hour days. Naveen has shown it’s possible to be multi-faceted, and lives all four of his distinct titles with excellence: Father, Husband, CEO, and Professor.
If he and his co-founders keep attracting the right people, not just because of their technological breakthroughs but also for their culture, they'll define the next era of AI at the Edge, and I’m betting they will.
#semiconductorindustry#artificialintelligence#analogcompute
EnCharge AI has received the 2025 Frost & Sullivan Technology Trailblazer award!
The award recognizes companies with an innovative and disruptive technology that can transform the industry while delivering substantial benefits and value to customers.
"EnCharge AI is catalyzing the shift from cloud-dependent AI to true on-device intelligence," said Manuel Albornoz, Best Practices Analyst at Frost & Sullivan. "The company's first-of-its-kind analog in-memory computing architecture is delivering breakthrough efficiency gains of 20x over traditional digital solutions, bringing advanced AI inference out of data centers and directly to edge devices for the first time. This is exactly the technology needed to unlock an entirely new generation of personalized and secure AI applications."
Learn more about how we’re building the future of local AI in the comments ⤵️
“To make our business case work, we had to really think down to the very fundamental levels of the laws of physics.”
Naveen Verma appeared on the From Startup to Grown Up Podcast to chat with host Alisa Cohn. In their conversation, Naveen and Alisa cover not only the challenges of a founder building a business but also the challenges of working in deep tech and grappling with the laws of nature themselves.
Tune in for the fascinating journey of EnCharge AI, from being just a technical possibility to a fast-growing company that’s set to transform AI chip efficiency and the entire AI industry as a result.
Find the link to the podcast in the comments!
Great to see CRN featuring EnCharge AI in its coverage of the industry shift toward neural processing units (NPUs), enabling PCs to run AI and machine learning workloads without the energy demands of traditional GPUs.
“Santa Clara, Calif.-based startup [EnCharge AI] announced back in May a 200-TOPS NPU that can use as little as 8.25 watts in an M.2 form factor for laptops as well as a four-NPU PCIe card that can provide roughly 1,000 TOPs to provide what it called ‘GPU-level compute capacity at the fraction of the cost and power consumption.’”
Read more:
https://lnkd.in/edSpyRa4
Our CEO Naveen Verma kicked off Season 10 of the Aerospace & Defense Technology podcast, discussing EnCharge’s debut AI accelerator, EN100.
EN100 is designed for the unique demands of future AI-driven aerospace and defense platforms, addressing the critical efficiency and reliability gaps traditional chips can’t.
Listen to the full episode here:
https://lnkd.in/et3Y7Yv2
EnCharge AI has been selected for EE Times | Electronic Engineering Times prestigious 2025 Silicon 100 list!
After evaluating companies across technology readiness, intellectual property, strategic partnerships, scalability, and market potential, the Silicon 100 places us among the industry's most promising startups in the Data Centers, AI Processors, and Networking category.
We believe the next generation of AI applications will require transformative solutions that bring compute closer to users, moving beyond data centers to unlock AI's full potential across all use cases. This recognition reaffirms that many in the industry share our vision, and we're energized by the momentum behind what we're building alongside our partners.
Thank you to EE Times!
Link to the full 2025 Silicon 100 in the comments.