Multiverse Computing Tech Stack

AI model compression using quantum-inspired techniques for cost and energy efficiency

Software Development San Sebastián, Basque Country 201–500 employees Founded 2019 Privately Held

Multiverse Computing builds compressed AI models derived from quantum computing principles, reducing inference costs and energy consumption by over 50% without performance loss. The hiring mix—sales-heavy with concurrent engineering build—suggests a company transitioning from R&D into market execution, with active projects spanning Fortune 500 deployments, sector-specific GTM strategies (pharma, financial services, public sector), and LLM pipeline infrastructure. Stack choice (PyTorch, TensorFlow, vLLM, DeepSpeed, Hugging Face) confirms focus on large language model optimization; adoption of Singularity containerization and SIEM indicate scaling toward production reliability and security.

Tech Stack 71 technologies

Core StackPython PyTorch NumPy Pandas Docker TensorFlow AWS Kubernetes Linux Slack Fortinet scikit-learn Airtable Monday.com Looker Tableau Power BI Hadoop Apache Spark Singularity GCP Azure macOS Microsoft 365 JAMF Hugging Face Transformers GitOps vLLM DeepSpeed SQL+40 more

AdoptingSingularity SIEM

What Multiverse Computing Is Building

◆Challenges

Reducing inference costs
Roi
Sovereignty
Reliability, robustness, and performance
Compressing large language models
Inference cost
Ensuring communication consistency
High-severity incident management
Optimizing trade-offs between accuracy and latency
Digitalization phase

▲Active Projects

Deploy cutting-edge ml/llm models to fortune global 500 clients
Design and implement ml/llm pipelines
Establish monitoring and alerting systems for llm performance
Poc development for compactifai and singularity
Gtm strategy for pharma & healthcare
Public sector go-to-market strategy
Customer feedback loop for product roadmap
Gtm sectoral strategy for fsi and retail
Poc and demo coordination
Llm compression using quantum-inspired technologies

Hiring Activity

Decelerating65 roles · 20 in 30d

Department

Sales

Engineering

Ops

Support

Marketing

Finance

Seniority

Senior

Mid

Manager

Director

Intern

Junior

Notable leadership hires: Engineering Director

Company intelligence

Find more companies like Multiverse Computing by tech stack, pain points and active projects

Get started free

About Multiverse Computing

Multiverse Computing is an AI model provider specializing in compression techniques that reduce the computational footprint of large language models and other AI systems. Founded in 2019 and headquartered in San Sebastián, Spain, the company operates offices across Canada, the United Kingdom, France, Germany, Italy, and the United States. Their core product applies quantum-inspired algorithms to shrink model size and inference cost while maintaining accuracy. The company sells to multinational enterprises across financial services, energy, telecommunications, and pharma sectors. Current operational focus spans model deployment pipelines, customer feedback integration, and go-to-market expansion across vertical and geographic segments.

HeadquartersSan Sebastián, Basque Country

Company Size201–500 employees

Founded2019

Hiring MarketsSpain, Canada, Germany, France, Italy, Philippines, United States, United Kingdom

Frequently Asked Questions

What is Multiverse Computing's core technology?

AI model compression using quantum-inspired techniques. Their approach reduces inference cost and energy consumption by more than 50% compared to uncompressed models without sacrificing performance.

What tech stack does Multiverse Computing use?

Python, PyTorch, TensorFlow, Hugging Face Transformers, vLLM, DeepSpeed, Docker, Kubernetes, AWS/GCP/Azure. They are actively adopting Singularity containerization and SIEM tooling.

Where is Multiverse Computing hiring?

Spain, Canada, Germany, France, Italy, Philippines, United States, United Kingdom, Switzerland, Belgium, and Poland. The company has 66 active roles with emphasis on sales and engineering positions.