EnCharge AI

By 2027, Gartner predicts that 40% of AI data centers will face operational constraints due to power availability. This isn’t just a technological roadblock—it’s a critical challenge that could slow AI’s growth and its broader potential. AI's growing energy demands are already reshaping the U.S. power grid. Data centers, especially those supporting AI workloads, are consuming vast amounts of electricity—sometimes more than entire countries. As the energy needs of AI continue to expand, regions like Northern Virginia, home to dense clusters of data centers, are hitting capacity limits. Utilities are scrambling to meet this surging demand, which is putting significant strain on existing infrastructure. Traditional computing systems are nearing their limits. The von Neumann bottleneck, where data constantly moves between memory and processors, hinders computational efficiency and escalates energy consumption. Analog in-memory computing offers a solution, performing calculations directly within memory arrays and cutting out the energy-intensive data movement that slows performance. This can deliver up to 20 times the efficiency of traditional systems. We’re creating an energy-efficient computing platform capable of driving sophisticated AI systems without the massive power draw that is becoming unsustainable. Our solutions offer precision, reliability, and scalability, outperforming digital solutions while solving decades-old problems that have made analog solutions previously infeasible. AI’s future isn’t about consuming more power; it’s about using energy smarter. By creating more efficient, sustainable computing systems, EnCharge is helping ensure AI’s growth can continue without outstripping the grid’s ability to power it. 

EnCharge AI proudly welcomes Jason Huang and Leslie Szeto, MSOD to our leadership team! Following our recent $100M Series B, Huang is joining as VP of Finance, and Leslie as Director of HR, to help scale our operations as we transition from development to commercialization of our breakthrough AI accelerators. Jason brings decades of experience steering tech companies through growth milestones, including leading SupplyShift through its recent acquisition and Arteris through its successful IPO. Leslie comes with 15+ years as an HR strategic leader at companies like Box, Intel, and Adobe. These key appointments, along with last month's addition of Dr. Shwetank Kumar as Chief Scientist, strengthen our organizational foundation. Together, they position us to bring our analog in-memory computing technology to market, delivering breakthrough compute efficiency for AI from edge to cloud. Please join us in welcoming Jason Huang and Leslie Szeto to EnCharge! https://lnkd.in/e9-TyuWa

The world is standing at the edge of an AI innovation gap. Breakthroughs in climate science, drug discovery, and personalized medicine are on the horizon if the industry can deliver the compute capacity that will support continued AI model development. However, the "memory wall" associated with conventional compute architectures – where processing and memory units operate separately – creates a debilitating bottleneck, forcing continuous data movement that consumes excessive energy, increases costs, and restricts AI deployment to well-resourced data centers with massive power budgets. Ultra-efficient AI processing represents a paradigm shift by integrating computation and memory into unified architectures. By performing calculations directly where data resides, these systems eliminate the bottleneck that has constrained AI's potential in many contexts. This approach dramatically reduces energy consumption by up to 20x, increases compute density 9x, and will democratize access to powerful computational tools across industries and applications, from edge to cloud. Ultra-efficient AI processing will help empower AI to solve critical problems while significantly reducing costs, energy consumption, and carbon footprint. 

Over the last few weeks two aspects have emerged 1) demand for AI Inference is exploding, 2) AI Inference is melting GPUs (thanks to the Ghibli craze!). At EnCharge AI, we're tackling this head-on by building energy-efficient inference accelerators. We're hiring! I'm looking for passionate minds across the SW stack—from firmware to LLM deployment, including quantization, compilers, and runtimes. Check out our openings: https://lnkd.in/g2efQjsW We have multiple openings for each of these roles and are flexible on job locations within US, Canada and parts of Europe. Whether you're just starting out or a seasoned pro, let’s connect! #ai #inference #compiler #runtime #llms

This year, the U.S. will see a shift in military operations from centralized to edge-based decision-making. Modern autonomous systems often require instant processing capabilities in diverse environments for efficient and responsive operations. As a result, edge computing is becoming essential in modern defense infrastructure, enabling smart decision-making in the most remote environments. This integration of AI at the tactical edge allows military personnel to receive AI-driven insights, informing human choices in dynamic situations. Edge computing also brings important advantages for data security in defense applications. By processing information closer to its source, military operations can enhance their security posture while maintaining operational efficiency. This approach naturally complements existing security frameworks while supporting mission performance. As a participant in the Defense Advanced Research Projects Agency (DARPA) OPTIMA program and with backing from RTX, we're working to enable advanced AI to run efficiently on edge devices rather than being confined to data centers. This edge-based approach is key to making military operations more efficient, secure, and responsive.

Join our CEO Dr. Naveen Verma for a panel discussion hosted by Normal Computing on "Technology Innovations in Next-Gen Computing, Infrastructure, and Algorithms to Tackle the AI Energy and Silicon Complexity Crises" on April 3rd! As AI systems continue to consume exponentially more energy, the semiconductor industry faces unprecedented challenges in meeting computational demands while maintaining sustainability. This exclusive workshop on “Solving the AI Energy and Silicon Complexity Crises with Physics-Based Computing” will bring together leading voices from research, industry, and government to explore physics-based computing approaches that could reduce AI training energy consumption by 1000x. Click the link to learn more and register: https://lu.ma/oi711a3v 

AI's energy consumption is rapidly becoming unsustainable. Data centers are projected to consume 500 terawatt-hours annually by 2027—equivalent to France's entire electricity usage. This isn't just an infrastructure challenge; it's an existential threat to AI advancement.   This is why Anzu Partners is proud to be early investors in EnCharge AI, where the team is pioneering a fundamental solution through our analog in-memory computing technology, which reduces energy needs by 10-20x. Their approach based on decades of research at Princeton University doesn't just shorten the distance between data and processing—it reimagines the entire compute architecture.   What currently requires a 1GW power plant could potentially run on just 100MW with our technology. This isn't incremental improvement; it's a paradigm shift that makes AI both powerful and sustainable.   This is why we're excited about our growing team, including our new Chief Scientist Dr. Shwetank Kumar, as they work to ensure AI's future isn't constrained by energy limitations.   Learn more: TerraWatt Timebomb: https://lnkd.in/g_NFqD6e Meet Dr. Shwetank Kumar: https://lnkd.in/gRVxWdQ7 #AI #Innovation #SustainableTech #EnergyEfficiency #GTC25 Naveen Verma, Kailash Gopalakrishnan, Echere Iroaga, Roman Mueller, Joseph F. Krause, Cathryn Paine, Jonathan Morris, Rohit Iragavarapu, John Kwaak, Siddharth Gupta, Nate Bender, Joon Lee, HoChan Lee, Scout Ventures, Philipp Schams, SIP Global Partners, John Tseng, Kai Tsang, Christy Chou, Silicon Catalyst Angels, Defense Advanced Research Projects Agency (DARPA)

Join our CEO, Dr. Naveen Verma at Normal Computing’s “Solving the AI Energy and Silicon Complexity Crises with Physics-Based Computing” event on April 3rd in NYC! Click here for the full agenda and registration link: https://lu.ma/oi711a3v

“We're here with agentic AI right now—an incredibly powerful technology that military and intelligence agencies should be implementing today, not tomorrow.” At last week's Tectonic Defense Summit, our COO, Echere Iroaga, shared how EnCharge is advancing edge AI solutions. Maintaining superiority in the AI era will require moving AI out of the data center and empowering military and intelligence personnel to leverage real-time insights directly on devices in contested environments like forward bases and war zones. With ultra-efficient AI chips capable of running advanced AI models securely on power-constrained devices like phones, laptops, and drones, EnCharge is bringing computational power to the tactical edge. We're proud to be part of an $18.6M Defense Advanced Research Projects Agency (DARPA) grant and funding from RTX Ventures and IQT (In-Q-Tel), to advance next-generation AI capabilities essential for defense applications.

EnCharge AI proudly welcomes Dr. Shwetank Kumar to the Executive Team as our new Chief Scientist! Dr. Kumar brings exceptional experience in advanced AI models including LLMs and SLMs, with a distinguished career spanning technical and business leadership roles at Intel, IBM, Synaptics, Orbital Insight, and Turo. As the author of the insightful "AI Afterhours" Substack (which will continue alongside his work with us), he combines deep technical expertise with strategic vision. "Having known Shwetank since 2007, I couldn’t be more pleased that he’s joined us at EnCharge," said our CTO and Co-Founder Kailash Gopalakrishnan. "Shwetank has drawn on his exceptional background to become a true thought leader on the future of AI." As Chief Scientist, Dr. Kumar will help shape our research agenda, product roadmap and strategic direction as we continue our mission to deliver unparalleled energy-efficient AI inference solutions from edge to cloud. In Dr. Kumar's own words: "EnCharge AI is addressing one of the industry's most pressing concerns—managing AI's growing energy demands sustainably. I am thrilled to join this exceptional team and help deliver the computational efficiency breakthroughs needed for AI's future." Please join us in welcoming Dr. Kumar to EnCharge! https://lnkd.in/efNUwPp9