EnCharge AI
Component
Analog in-memory computing AI accelerators that cut inference power, cost and CO2 versus GPUs.
EnCharge AI builds analog in-memory computing accelerators - chips that compute where the memory lives instead of shuttling data between memory and processor. The company reports 20x higher efficiency (TOPS/W), 9x higher compute density (TOPS/mm2), 10x lower total cost per inference and about 100x lower CO2 emissions versus cloud, measured on silicon against industry-leading products, plus 350M+ chips shipped and 150+ patents granted (all company-reported). The core technology came out of six years of Princeton University research initially funded by DARPA and the U.S. Department of Defense, and is designed to drop into the existing semiconductor supply chain as chiplets, ASICs or standard PCIe cards with an end-to-end programmable software stack. EnCharge closed an oversubscribed $100M Series B in February 2025, bringing total funding past $144M.
Research mapping
Physical AI / Edge Compute & Control Silicon (L5) / Component
Evidence boundary
Original analysis and independent corroboration remain under development. Commercial scale and durable technical advantage remain under review.
What makes it defensible
Charge-based analog in-memory compute silicon with 150+ granted patents and DARPA/DoD lineage - the efficiency edge comes from device physics, not just chip architecture.
Position in the stack: Component supplier of AI accelerators that sit inside edge and cloud systems. Defensibility comes from an in-memory computing macro that is difficult to replicate and a patent portfolio validated through U.S. government research programmes.
Who buys it
Founding team
Naveen Verma — Co-founder and CEO — Professor of electrical and computer engineering at Princeton University since 2009; spent six years leading the research behind EnCharge's core technology before co-founding the company
Kailash Gopalakrishnan — Co-founder and CTO — Former IBM Fellow who led worldwide efforts on advanced AI hardware, software and algorithmic co-design at IBM
Echere Iroaga — Co-founder and COO — 25+ years in semiconductor design and management; previously VP/GM of MACOM's connectivity business unit and founder of Renovus Inc.
SourceFinancing detail
- Last round
- Series B
Mapped funds and investors
Direct relationships supported by published evidence.
Reported holdings as-of
No sourced relationships mapped yet.
Disclosed investment history
No sourced relationships mapped yet.
Showing up to four relationships in each group. ETF weights describe the fund’s portfolio, not ownership of this company. Private investment history does not establish current ownership.
Evidence & media
-
EnCharge AI raises over $100 million in funding to bring AI inference chips to market
What this source supports: >$100M Series B led by Tiger Global to commercialize analog in-memory-computing AI chips.
What it does not establish: No revenue, customer commitments, or production volumes.
Source -
EnCharge AI launches its analog in-memory EN100 AI accelerator
What this source supports: EN100 analog in-memory AI accelerator for laptops/edge, claimed up to 20x efficiency.
What it does not establish: Vendor performance claims; not independently benchmarked.
Source -
In-Memory Computing Chip Is a Processing Breakthrough for On-Device AI Applications
What this source supports: Third-party coverage of EnCharge's analog in-memory computing chip.
What it does not establish: Explainer; no independent performance measurements.
Source
Timeline
-
Financing
Series B — $144,000,000 total disclosed
Source