The partnership pairs LeapThought's own AI with Google Cloud's Gemini models and sovereign Cloud offering, giving infrastructure owners control of their data and connected intelligence across the full project and asset lifecycle.
SINGAPORE – Media OutReach Newswire – 23 October 2026 – LeapThought and Google Cloud today announced a strategic partnership to bring sovereign, domain-native AI to the owners and operators of critical building infrastructure, giving them control of their data and sharper decision-making across the full value chain, from delivery through operations.
Ratnakar Garikipati, Founder, Chair, and CEO of LeapThought, alongside Rahul Sharma, Google Cloud Managing Director, APAC Public Sector. The partnership combines LeapThought's specialized AI with Google Cloud's Gemini models and sovereign infrastructure to deliver secure data control and end-to-end intelligence for infrastructure assets.
The partnership has two priorities. The first is value chain integration, which connects everyone involved in delivering and running an asset — owners, operators, developers, consultants, contractors, and suppliers — on one governed platform. The second is security and sovereignty: owners can keep control of their data, within compliant, standardized frameworks that can be repeated from one program to the next.
FULCRUMHQ, LeapThought's globally recognized Integrated Digital Delivery (IDD) platform, gives infrastructure owners the option to run on Google Cloud, including Google's sovereign, air-gapped solutions for defense, public sector, and private sector environments requiring high security. At its core is GENIEHQ, LeapThought's own Industrial AI, embedded directly into the platform and built to reason over the structured and unstructured data that defines a capital program — BIM models, GIS and geospatial data, ERP financials, schedules, contracts, reality capture, telemetry, and data from field robotics and sensors. The partnership pairs GENIEHQ with Google Cloud's Gemini models, giving owners and their delivery teams natural-language access to that data across a project and a portfolio.
The partnership delivers on these two priorities as follows:
1. Value Chain Integration: One Governed Record, Natural-Language Insight
Major capital and building infrastructure programs span owners, operators, developers, consultants, contractors, and suppliers — each typically working in separate systems, with design, construction, and operational data fragmented across handovers. LeapThought's FULCRUMHQ platform brings this information together into a single governed record that follows an asset from design through operation, giving owners a consistent view across their portfolio, not just a single project.
By running FULCRUMHQ — and the GENIEHQ Industrial AI embedded within it — on Google Cloud, and integrating LeapThought's proprietary data architecture with Google's Gemini models, the partnership enables:
- Interactive Model and Portfolio Querying: Combining GENIEHQ's deep understanding of project and portfolio data with the multimodal capabilities of Gemini, owners, engineers, project managers, and field teams can ask complex, plain-language questions directly against BIM models and portfolio data — instead of manually clicking through complex 3D CAD/BIM model trees or running predefined filter scripts — such as identifying clash detections, extracting exact material quantities, verifying regulatory code compliance, or surfacing risk trends across a program (e.g., "Show me all fire-rated doors on Level 3 that don't match the latest spec," or "Which contractors across our portfolio have the highest rate of non- conforming work this quarter").
- Global Scale and Reliability: Google Cloud provides the high-performance compute and elastic scalability needed to process compute-intense multi-gigabyte 3D models and point-cloud datasets, across a single project or an entire portfolio, in real time.
2. Security and Sovereignty: Full Control on Google Cloud's Sovereign Solutions
For critical civil assets, defense installations, public sector agencies, and private sector owners where data cannot traverse the public cloud, LeapThought gives owners the choice to deploy on Google Cloud's data residency solutions, including Google Cloud Data Boundary, Google Cloud Dedicated, and Google Distributed Cloud. In these environments, GENIEHQ, LeapThought's Industrial AI embedded within FULCRUMHQ, runs alongside Gemini locally, within the owner's chosen security boundary, ensuring zero cross-boundary exposure or unauthorized external telemetry.
"The value of an asset now lies as much in the intelligence around it as in its physical form. That intelligence belongs to the owner, and it must be secure, sovereign, and shared across the whole value chain. Demand for this is growing quickly, in Asia and around the world, and that is what brought us and Google Cloud together. We're starting from Singapore and building toward a standard for critical infrastructure everywhere," said Ratnakar Garikipati, Founder, Chairman, and CEO of LeapThought.
"Public sector and critical infrastructure agencies manage some of the most intricate engineering and regulatory datasets in the world. The partnership between Google Cloud and LeapThought enables project teams and portfolio owners to quickly surface actionable insights directly from complex BIM models and program data using everyday language. Combined with Google's sovereign, air-gapped solutions, we're providing agencies with the option to modernize project workflows while maintaining strict data governance, residency, and operational sovereignty," said Rahul Sharma, Managing Director, APAC Public Sector, Google Cloud.
Hashtag: #LeapThought
The issuer is solely responsible for the content of this announcement.
About LeapThought
LeapThought is a US technology company that builds software connecting people, data, and AI across the infrastructure lifecycle. It is headquartered in Bellevue, Washington, has its Asia headquarters in Singapore, and operates across the United States, Brazil, Singapore, and New Zealand. Its FULCRUMHQ platform and embedded GENIEHQ Industrial AI serve public agencies, developers, consultants, contractors, and universities and institutes of higher learning. Clients include Singapore's Land Transport Authority, Building and Construction Authority, and JTC. The company was founded and led by Ratnakar Garikipati.
** This press release is distributed by Media OutReach Newswire through automated distribution system, for which the client assumes full responsibility. **
New hub connects North American AI technology and demand with Southeast Asia’s expanding infrastructure ecosystem; company is building capacity across North America, Asia-Pacific and Europe
SAN FRANCISCO, US – Media OutReach Newswire – 23 October 2026 – TokenPulse, a next-generation AI inference infrastructure company headquartered in San Francisco and Singapore, today announced the opening of its Asia-Pacific Operating Hub in Singapore. The company marked the occasion with an opening ceremony attended by invited industry leaders and ecosystem partners, who gathered to discuss emerging opportunities in AI compute and global infrastructure.
TokenPulse Opens Singapore APAC Operating Hub
Companies putting AI applications into production face rising inference costs, pressure to keep response times low, and often have to rebuild their platforms as workloads grow. Inference demand is climbing alongside: the global AI infrastructure market is projected to grow from $982 billion in 2025 to $1.98 trillion by 2027, doubling in size over the two-year period.
TokenPulse delivers high-performance, cost-efficient AI compute at global scale. By combining data-center resources, GPU infrastructure, an optimized inference engine and globally coordinated serving capabilities, the company serves foundation model companies, AI platforms, enterprises and AI-native applications.
Singapore as a Strategic Hub for a Global Infrastructure Network
TokenPulse is building its infrastructure footprint across North America, Asia-Pacific and Europe, with Singapore serving as its strategic center for APAC. The company's regional strategy is designed to connect North America's AI technology, customer and capital ecosystems with the expanding data-center and energy resources of Southeast Asia.
From Singapore, TokenPulse is positioned to work with partners across Malaysia, Indonesia, Thailand and the wider region. The hub will support regional operations, infrastructure partnerships and coordination across the company's global inference network, helping bring compute capacity closer to customers and AI workloads.
TokenPulse is taking a long-term approach to building an integrated AI infrastructure network, bringing together data-center, energy and compute resources to support growing AI inference demand across its global markets. The company is advancing multiple data-center projects with infrastructure partners across key markets, positioning itself for sustained capacity growth. Its infrastructure roadmap includes deployment of the latest-generation NVIDIA AI infrastructure, while its engineering focus remains on improving how effectively compute resources are converted into tokens and useful AI output.
"AI is undergoing a fundamental shift: compute demand is moving from training to inference. For businesses, inference is no longer a hidden cost sitting in the background. It is becoming a core competitive factor in determining whether AI products can successfully reach the market," said Steven Zhou, CEO of TokenPulse.
"As one of the fastest-growing regions for AI applications, developers in Asia Pacific are direct about cost and performance. By making Singapore our co-headquarters and APAC operating hub, we are putting our compute and our team closest to them to deliver better performance at lower cost."
Flexible Consumption and Deployment Options
TokenPulse offers three consumption models on a unified technology stack, allowing customers to start with smaller workloads and expand their compute capacity as their applications grow.
- Serverless — Pay per Token: Provides unified API access for rapid deployment, model evaluation and variable workloads. It enables developers and AI-native applications to get started without committing to dedicated GPU capacity upfront.
- On-Demand — Pay per GPU-Hour: Offers dedicated endpoints and predictable performance for customers with elastic compute requirements, supporting workload optimization as usage grows.
- Reserved — Contracted Capacity: Provides dedicated, contracted compute capacity for sustained workloads and more customized infrastructure requirements.
TokenPulse offers three deployment options to accommodate different infrastructure, security and operational requirements:
- TokenPulse Cloud. Managed GPU infrastructure for a streamlined path to production.
- Private VPC. A TokenPulse-hosted environment with a dedicated network boundary and workload isolation.
- BYOC (Bring Your Own Cloud). The TokenPulse inference engine can be deployed in a customer's cloud or VPC, or on the customer's own GPUs.
The company's commercial approach is built around a simple principle: "Start small. Scale without replatforming." Customers can begin by validating their applications through API-based access, then transition to GPU-based services or reserved capacity as demand increases. This lets businesses expand their AI workloads without having to switch providers or rebuild their platforms as their infrastructure needs change.
Optimizing Inference for Performance and Cost
TokenPulse's competitive advantage is the combination of high-standard data center infrastructure and deep inference optimization tailored to specific industry use cases. By optimizing open-weight models for vertical workloads, TokenPulse improves compute efficiency and serving throughput so customers can process more tokens at lower cost.
Key technology capabilities include:
- Higher cache efficiency: TokenPulse uses HiCache and a multi-tier key-value (KV) caching architecture to reduce repeated computation. Its observed cluster-level cache hit rate can reach 95%.
- Faster response times: With context parallelism enabled, TokenPulse has achieved a 50% reduction in 90th-percentile time to first token (P90 TTFT) and a 75% reduction in 99th-percentile TTFT (P99 TTFT) compared with the same workload without context parallelism enabled.
- Global intelligent routing: A unified routing architecture directs requests across serving regions based on factors including customer requirements, model, region and time window. Nearest-edge access, session-context reuse and redundant network paths are designed to support low latency, high availability and efficient compute utilization.
Together, these capabilities are designed to help AI businesses improve inference economics while maintaining the responsiveness and reliability required for production workloads. Actual performance gains vary by workload and operating conditions.
Leadership With Deep Infrastructure Experience
TokenPulse's leadership and technical team bring experience spanning AI and machine learning infrastructure, cloud computing, networking, data centers and global business operations, including backgrounds at Microsoft, Google and Equinix.
CEO Steven Zhou is a serial entrepreneur with more than 20 years of experience scaling businesses across the United States and Asia. Chairman Thomas Pang brings more than 30 years of experience across government, investment, infrastructure and corporate leadership. He spent more than two decades with Keppel and served as CEO of Keppel's Data Centres & Networks Division from 2014 to 2024, overseeing the group's global data-center and digital-connectivity businesses.
"Singapore is a natural choice for building our APAC operating hub, given its mature data center ecosystem, strong network connectivity and deep talent pool," said Thomas Pang, Chairman of TokenPulse. "Our foundations are rooted in decades of experience across data centers, networking and cloud computing. Drawing on that experience in building and operating infrastructure at scale, we will continue investing in compute capacity and ecosystem support, delivering the best performance at the lowest cost per token."
Hashtag: #TokenPulse
The issuer is solely responsible for the content of this announcement.
About TokenPulse
TokenPulse is a next-generation AI inference infrastructure company headquartered in San Francisco and Singapore. The company combines latest-generation GPU infrastructure, an optimized inference engine and global AI infrastructure capabilities to deliver high-performance, cost-efficient compute for foundation model companies, AI platforms, enterprises and AI-native applications worldwide.
With flexible Serverless, On-Demand and Reserved consumption models, TokenPulse enables customers to start with API-based inference and scale to dedicated GPU capacity as their workloads grow — without replatforming.
Learn more: www.tokenpulse.com
** This press release is distributed by Media OutReach Newswire through automated distribution system, for which the client assumes full responsibility. **
