Skip to main content

Majestic Labs

Category: AI Infrastructure

Next-generation AI servers with 1000x the memory capacity of top GPUs, designed to solve the memory wall bottleneck in AI infrastructure Majestic Labs was founded in 2023. The company is led by Ofer Shacham. Based in San Francisco, USA & Tel Aviv, Israel. Team size: 11-50. Total funding raised: $100 million. Latest round: Series A, $80M, 2025-11-10, led by Bow Wave Capital. Key investors include Bow Wave Capital (Series A lead), Lux Capital (Seed lead), SBI, Upfront Ventures, Grove Ventures, Hetz Ventures, QP Ventures, Aidenlair Global, TAL Ventures.

Founded
2023
Headquarters
San Francisco, USA & Tel Aviv, Israel
Team size
11-50
Total funding
$100 million

Value proposition

To redefine AI infrastructure by overcoming the 'memory wall,' which is the growing disparity between the speed of processors and the ability of memory to supply them with data. This allows for the creation of a new class of AI models that are currently impossible to build.

Products and solutions

AI server that rebalances memory and compute, offering up to 1000x the memory capacity of current systems, All-in-one server architecture that consolidates multiple server racks into a single unit, Custom accelerator and memory-interface chips with up to 128TB of high-bandwidth memory per server

Unique value

Majestic Labs is re-architecting the entire AI server system, rather than just focusing on component-level improvements. Their approach integrates hardware and software to hyperscale memory capacity within familiar frameworks.

Target customer

Enterprises and organizations that require large-scale AI model training and inference, such as cloud service providers and AI-focused companies

Industries served

AI Infrastructure, Cloud Computing, Data Centers

Technology advantage

The company's key technological advantage is its ability to 'tear down the memory wall,' enabling a significant increase in memory capacity through custom silicon that disaggregates memory from compute. Their business advantage lies in its founding team of experienced executives from Google and Meta who have a deep understanding of the challenges in AI infrastructure. The founding team collectively holds over 120 patents from their previous work.

How they differentiate

Majestic Labs is developing next-generation servers that deliver 1000x the memory capacity of a top-of-the-line GPU for AI workloads, consolidating multiple server racks into a single unit to solve the 'memory wall' bottleneck in AI infrastructure.

Main competitors

Element Labs, Speedata, Apex

Key partnerships

Bow Wave Capital, Lux Capital, SBI, Upfront Ventures, Grove Ventures, Hetz Ventures, QP Ventures, Aidenlair Global, TAL Ventures

Notable customers

No publicly disclosed customers yet - company is in pilot deployment phase

Major milestones

Emerged from stealth with $100M in financing in November 2025, Founded by former executives from Google and Meta, Developing AI servers with 1000x memory capacity of GPUs, Team collectively holds over 120 patents

Growth metrics

The company has recently emerged from stealth, so public growth metrics are not yet available. They are currently in the pilot deployment phase with customers.

Market positioning

Positioned as a provider of high-performance AI infrastructure with a disruptive approach to system architecture, focusing on memory capacity to offer significant performance and efficiency gains.

Geographic focus

San Francisco, USA and Tel Aviv, Israel

Patents and IP

The founding team collectively holds over 120 patents in custom silicon and system design from their previous work at Google, Meta, and earlier ventures. Specific patents filed under Majestic Labs Inc. are not yet publicly disclosed.

About Ofer Shacham

Co-Founder & CEO at Majestic Labs. Previously VP, Head of Silicon at Meta's Reality Labs and Senior Director of Engineering at Google, where he led silicon teams for Pixel Visual Core and TPU projects.

Official website: