Tech and communications companies are both creators and heavy users of AI. From network optimization that ensures reliable connectivity, to content moderation at scale, to AI-powered customer support, these industries showcase cutting-edge AI applications.
Open a case for full evidence, sources, and implementation details.Click any column header to sort & filter — e.g. Market for industry, domain & country.
Alibaba Cloud's Tair KVCache team and storage hardware-software integration team upgraded the open-source 3FS file system to support enterprise KVCache storage for AI inference.,The work optimized RDMA load balancing and small I/O, added a user-space persistence engine, introduced GPU Direct RDMA and multi-tenant isolation, and built a Kubernetes Operator for one-click deployment, self-healing, elastic scaling, and monitoring.,The solution was integrated with SGLang, vLLM, and Tair KVCache Manager to improve long-context and agent-style inference performance.
Turbo AI is a technology company founded by experts in cloud computing, distributed systems, and machine learning.,It provides vertically integrated infrastructure services for organizations building, training, and running AI models.,The company partnered with Alibaba Cloud to solve multi-cloud complexity, large-scale data migration, and GPU cost and flexibility challenges.
Trend Micro enhanced its Companion AI assistant in its Vision One security software with long-term and short-term agentic memory capabilities.,The company used Amazon Neptune as a knowledge graph and experience layer to connect threat intelligence, alerts, investigations, and successful remediation patterns so analysts can get more explainable and personalized security recommendations.,Trend Micro combined Amazon Bedrock, Amazon Textract, and Amazon Managed Streaming for Apache Kafka to orchestrate agentic workflows, clean and enrich user conversations, and feed metadata into the graph for long-term memory and correlation analysis.
Innovation
4.3 / 5
Outcome
New product / capabilityBuilt long-term and short-term agentic memory
Druva built DruAI, an agentic AI solution to help customers complete cyber investigations and recovery workflows.,The solution uses a multi-agent system on AWS with AgentCore Memory, AgentCore Code Interpreter, Amazon Bedrock access to Claude models, voice capabilities with Amazon Nova Sonic, and Bedrock Guardrails.,In production, DruAI coordinates 8 to 10 specialized agents across investigation, help, and recovery workflows.
AnalyticDB Ray is a fully managed Ray service on Alibaba Cloud AnalyticDB for MySQL that supports cloud-native Data+AI workloads with hybrid CPU/GPU resource scheduling.,The article describes PB-level multimodal processing and distributed fine-tuning scenarios, including Qwen-VL multimodal model fine-tuning through LLamaFactory integrated into AnalyticDB Ray.
Kimi rolled out multiple agent-based products and built an end-to-end agent infrastructure with Alibaba Cloud to handle high-concurrency requests, fast sandbox startup, state preservation, secure isolation, and low-cost elastic scaling.,The architecture combines Alibaba Cloud Container Service for Kubernetes (ACK), ACS Agent Sandbox, Lindorm multi-model database, MicroVM isolation, NetworkPolicy, and Fluid to support production AI agents and model training workloads.
TELUS built its generative AI platform Fuel iX on Google Cloud to unify siloed data, increase trust in enterprise data, and support proactive problem-solving, customer support, and engineering productivity across the company.
C Spire built an agentic AI assistant on AWS to help network technicians solve issues more efficiently and keep its network resilient.,The solution gives technicians natural-language access to equipment manuals, technical documentation, and network schematics while combining alarms, logs, and configurations to support diagnosis and resolution.
Ericsson describes an AWS-hosted rApp as a Service solution that uses agentic AI to automate RAN operations and optimization for communications service providers.,The platform integrates with Ericsson Intelligent Automation Platform and O-RAN interfaces to coordinate specialized agents for anomaly detection, root-cause explainability, cell shaping, and uplink performance optimization.,It is delivered as a multi-tenant SaaS offering through AWS Marketplace.
Fujitsu faced time-consuming and manual sales proposal creation processes, hampering staff productivity and slowing the onboarding of new salespeople. To address these challenges, Fujitsu developed an orchestrated multi-agent AI system using Microsoft Azure AI Agent Service, Azure AI Foundry, and Semantic Kernel, allowing them to automate data gathering and proposal drafting. The new system integrates with Fujitsu's existing workflows and leverages multiple specialized AI agents to dynamically retrieve, synthesize, and tailor sales proposals from dispersed enterprise data sources. As a result, Fujitsu boosted sales team productivity by 67%, enabling staff to focus on customer engagement, faster onboarding for new hires, and delivery of more accurate, data-driven, and current proposals.
ContraForce is a cybersecurity startup that built a multi-tenant Security Delivery Platform for managed security service providers (MSPs).,The platform operationalizes Microsoft Sentinel, Microsoft Defender XDR, Microsoft Entra ID, and Microsoft Foundry / Azure OpenAI, with a no-code automation layer called Gamebooks.,AI agents triage, investigate, and execute supervised or autonomous response actions so MSPs can scale 24/7 SOC operations and manage more customers.
Microsoft Build 2025 showcased the company's comprehensive strategy for developing the 'agentic web'—an ecosystem centered around the deployment of autonomous and multi-agent AI systems. Key announcements included Azure AI Foundry for multi-agent orchestration, next-generation GitHub Copilot coding agents capable of autonomous code refactoring and defect fixing, and Windows AI Foundry for local, on-device AI execution. Microsoft introduced new protocols for agent interoperability (A2A, MCP), enterprise-grade governance (Entra Agent ID), and advanced governance with Microsoft Purview integration. The standout real-world application, Microsoft Discovery, demonstrated the acceleration of scientific R&D—screening over 367,000 materials in 200 hours to rapidly develop a non-PFAS immersion coolant for data centers. The article illustrates a production-ready shift towards autonomous and multi-agent AI solutions, impacting software development, scientific research, and enterprise workflow automation.
Microsoft announced new Copilot Tuning and multi-agent orchestration in Microsoft 365 Copilot Studio, empowering organizations to build and fine-tune AI agents using domain-specific data, workflows, and processes. The low-code tools let non-data scientists create, deploy, and rapidly update custom AI agents. The release includes multi-agent orchestration, allowing agents to collaborate on complex cross-departmental workflows, such as onboarding and support, and integration with Azure AI Foundry so users can bring their models. Noteworthy customer examples: Wells Fargo reduced bankers' search time by 95% (10 min to 30 sec) via an agent serving 35,000 staff; T-Mobile uses agents to aggregate product data from over 20 sources instantly; HCLTech accelerated case resolutions by 40% and redeployed 30% of a 500-strong support team. Security and compliance are ensured via Microsoft Entra Agent ID and Microsoft Purview Information Protection. The ecosystem includes an Agent Store for discovery, a Teams AI library for multi-agent communications, and pro-code/low-code developer tools.
Replit adopted Anthropic's Claude 3.5 Sonnet on Vertex AI to power Replit Agent, which turns natural-language prompts into working applications by handling environment setup, code generation and editing, testing, and deployment.,Replit also uses Gemini 1.5 Flash for additional AI features and runs supporting infrastructure on Google Cloud including Cloud Run, Compute Engine, Cloud SQL, and BigQuery.
Innovation
4.1 / 5
Outcome
Developers supported: More than 35,000,000 developers
Quench.ai used Google Cloud to build a scalable, cost-effective AI platform for employees to search across company tools and data in seconds.,The company used Gemini 2.5 with Google AI Studio, Cloud Run, Cloud SQL, Memorystore, and Identity-Aware Proxy to prototype and deploy a custom debugging tool and other AI workflows.,Quench.ai says the platform enables faster access to internal information and supports agentic AI development for onboarding and handover scenarios.
Innovation
4.1 / 5
Outcome
Speed & agilityAccelerated debugging tool development and deployment
Box builds customizable AI and AI agents with Google Cloud to democratize AI into workers' hands.,Box Agent and Box AI Studio let non-technical users create custom, purpose-built AI agents over Box content.,Agents authenticate to a company's Box instance and operate within existing user permissions and security settings.,Multi-agent systems use Agent2Agent (A2A) and Box MCP server to interact with other Gemini Enterprise agents.,BigQuery is used for reporting and processing structured aspects of unstructured data.
Innovation
4.1 / 5
Outcome
New product / capabilityEnabled no-code custom AI agents
Adya is a tech company with two primary business segments: Adya, a SaaS-based data security platform that protects data in cloud applications, and Adya.ai, an AI-powered platform that transforms enterprise workflows into AI-powered agents and copilots, improving efficiency and driving cost savings.,The company built a platform on Google Cloud that combines Gemini, Vertex AI, Google Agentspace, Google Kubernetes Engine, Pub/Sub, Cloud Storage, and Virtual Private Cloud to support secure enterprise AI workflows.,The platform includes App Studio and Model Studio for AI-assisted application generation, fine-tuned LLM development, and configurable deployment options for enterprise customers.
Enterprises have long relied on Microsoft Entra’s Conditional Access policies to safeguard organizational resources, but manually managing these policies has proven complex and error-prone due to overlapping rules, legacy authentication risks, and rapid environmental changes. The Conditional Access Optimization Agent (CAOA), powered by Security Copilot and integrated within Microsoft Entra, addresses these challenges by continuously scanning user and app inventories, sign-in logs, and existing policies to identify unprotected users, risky applications, and policy gaps. The AI-driven agent analyzes data against Zero Trust principles and provides actionable recommendations with one-click remediation, reducing manual workload while empowering administrators with full oversight. Pilot organizations have already reported freeing up security team hours and discovering over 900 unprotected users not identified through manual audits. CAOA complements existing tools—such as the What If tool and Gap Analyzer—by automating policy reviews while maintaining robust audit trails for transparency. The tool requires Microsoft Entra ID P1+ and Security Copilot licenses and currently operates in private preview. Through proactive, AI-assisted policy management, organizations can both strengthen security and reduce operational overhead.
IBM faced major productivity drag across HR, procurement, IT, supply chain, sales, and tax because employees spent time searching for policies, navigating siloed systems, handling routine support, and manually aggregating data.,IBM set out to create an enterprise-wide productivity blueprint using AI, hybrid cloud, and automation, with domain assistants such as AskIT, AskHR, and AskSales deployed across workflows.
Ericsson and AWS collaborate to deliver cloud-native, AI-powered modernized OSS/BSS for autonomous networks.,Ericsson embeds more than 20 cloud-native Telco IT AI Apps across OSS/BSS and uses Amazon SageMaker, Amazon Bedrock, Amazon Bedrock AgentCore, Amazon Q, and Kiro for generative and agentic workflows.,The article describes an Ericsson Telco Agentic AI Studio built on Amazon Bedrock, plus a Gen-AI Lab that helps fast-track ideas from experimentation to deployable products.
Alibaba Cloud's Tair KVCache team and storage hardware-software integration team upgraded the open-source 3FS file system to support enterprise KVCache storage for AI inference.,The work optimized RDMA load balancing and small I/O, added a user-space persistence engine, introduced GPU Direct RDMA and multi-tenant isolation, and built a Kubernetes Operator for one-click deployment, self-healing, elastic scaling, and monitoring.,The solution was integrated with SGLang, vLLM, and Tair KVCache Manager to improve long-context and agent-style inference performance.
Turbo AI is a technology company founded by experts in cloud computing, distributed systems, and machine learning.,It provides vertically integrated infrastructure services for organizations building, training, and running AI models.,The company partnered with Alibaba Cloud to solve multi-cloud complexity, large-scale data migration, and GPU cost and flexibility challenges.
Trend Micro enhanced its Companion AI assistant in its Vision One security software with long-term and short-term agentic memory capabilities.,The company used Amazon Neptune as a knowledge graph and experience layer to connect threat intelligence, alerts, investigations, and successful remediation patterns so analysts can get more explainable and personalized security recommendations.,Trend Micro combined Amazon Bedrock, Amazon Textract, and Amazon Managed Streaming for Apache Kafka to orchestrate agentic workflows, clean and enrich user conversations, and feed metadata into the graph for long-term memory and correlation analysis.
Druva built DruAI, an agentic AI solution to help customers complete cyber investigations and recovery workflows.,The solution uses a multi-agent system on AWS with AgentCore Memory, AgentCore Code Interpreter, Amazon Bedrock access to Claude models, voice capabilities with Amazon Nova Sonic, and Bedrock Guardrails.,In production, DruAI coordinates 8 to 10 specialized agents across investigation, help, and recovery workflows.
AnalyticDB Ray is a fully managed Ray service on Alibaba Cloud AnalyticDB for MySQL that supports cloud-native Data+AI workloads with hybrid CPU/GPU resource scheduling.,The article describes PB-level multimodal processing and distributed fine-tuning scenarios, including Qwen-VL multimodal model fine-tuning through LLamaFactory integrated into AnalyticDB Ray.
Kimi rolled out multiple agent-based products and built an end-to-end agent infrastructure with Alibaba Cloud to handle high-concurrency requests, fast sandbox startup, state preservation, secure isolation, and low-cost elastic scaling.,The architecture combines Alibaba Cloud Container Service for Kubernetes (ACK), ACS Agent Sandbox, Lindorm multi-model database, MicroVM isolation, NetworkPolicy, and Fluid to support production AI agents and model training workloads.
TELUS built its generative AI platform Fuel iX on Google Cloud to unify siloed data, increase trust in enterprise data, and support proactive problem-solving, customer support, and engineering productivity across the company.
C Spire built an agentic AI assistant on AWS to help network technicians solve issues more efficiently and keep its network resilient.,The solution gives technicians natural-language access to equipment manuals, technical documentation, and network schematics while combining alarms, logs, and configurations to support diagnosis and resolution.
Ericsson describes an AWS-hosted rApp as a Service solution that uses agentic AI to automate RAN operations and optimization for communications service providers.,The platform integrates with Ericsson Intelligent Automation Platform and O-RAN interfaces to coordinate specialized agents for anomaly detection, root-cause explainability, cell shaping, and uplink performance optimization.,It is delivered as a multi-tenant SaaS offering through AWS Marketplace.
Fujitsu faced time-consuming and manual sales proposal creation processes, hampering staff productivity and slowing the onboarding of new salespeople. To address these challenges, Fujitsu developed an orchestrated multi-agent AI system using Microsoft Azure AI Agent Service, Azure AI Foundry, and Semantic Kernel, allowing them to automate data gathering and proposal drafting. The new system integrates with Fujitsu's existing workflows and leverages multiple specialized AI agents to dynamically retrieve, synthesize, and tailor sales proposals from dispersed enterprise data sources. As a result, Fujitsu boosted sales team productivity by 67%, enabling staff to focus on customer engagement, faster onboarding for new hires, and delivery of more accurate, data-driven, and current proposals.
ContraForce is a cybersecurity startup that built a multi-tenant Security Delivery Platform for managed security service providers (MSPs).,The platform operationalizes Microsoft Sentinel, Microsoft Defender XDR, Microsoft Entra ID, and Microsoft Foundry / Azure OpenAI, with a no-code automation layer called Gamebooks.,AI agents triage, investigate, and execute supervised or autonomous response actions so MSPs can scale 24/7 SOC operations and manage more customers.
Microsoft Build 2025 showcased the company's comprehensive strategy for developing the 'agentic web'—an ecosystem centered around the deployment of autonomous and multi-agent AI systems. Key announcements included Azure AI Foundry for multi-agent orchestration, next-generation GitHub Copilot coding agents capable of autonomous code refactoring and defect fixing, and Windows AI Foundry for local, on-device AI execution. Microsoft introduced new protocols for agent interoperability (A2A, MCP), enterprise-grade governance (Entra Agent ID), and advanced governance with Microsoft Purview integration. The standout real-world application, Microsoft Discovery, demonstrated the acceleration of scientific R&D—screening over 367,000 materials in 200 hours to rapidly develop a non-PFAS immersion coolant for data centers. The article illustrates a production-ready shift towards autonomous and multi-agent AI solutions, impacting software development, scientific research, and enterprise workflow automation.
Microsoft announced new Copilot Tuning and multi-agent orchestration in Microsoft 365 Copilot Studio, empowering organizations to build and fine-tune AI agents using domain-specific data, workflows, and processes. The low-code tools let non-data scientists create, deploy, and rapidly update custom AI agents. The release includes multi-agent orchestration, allowing agents to collaborate on complex cross-departmental workflows, such as onboarding and support, and integration with Azure AI Foundry so users can bring their models. Noteworthy customer examples: Wells Fargo reduced bankers' search time by 95% (10 min to 30 sec) via an agent serving 35,000 staff; T-Mobile uses agents to aggregate product data from over 20 sources instantly; HCLTech accelerated case resolutions by 40% and redeployed 30% of a 500-strong support team. Security and compliance are ensured via Microsoft Entra Agent ID and Microsoft Purview Information Protection. The ecosystem includes an Agent Store for discovery, a Teams AI library for multi-agent communications, and pro-code/low-code developer tools.
Replit adopted Anthropic's Claude 3.5 Sonnet on Vertex AI to power Replit Agent, which turns natural-language prompts into working applications by handling environment setup, code generation and editing, testing, and deployment.,Replit also uses Gemini 1.5 Flash for additional AI features and runs supporting infrastructure on Google Cloud including Cloud Run, Compute Engine, Cloud SQL, and BigQuery.
Quench.ai used Google Cloud to build a scalable, cost-effective AI platform for employees to search across company tools and data in seconds.,The company used Gemini 2.5 with Google AI Studio, Cloud Run, Cloud SQL, Memorystore, and Identity-Aware Proxy to prototype and deploy a custom debugging tool and other AI workflows.,Quench.ai says the platform enables faster access to internal information and supports agentic AI development for onboarding and handover scenarios.
Box builds customizable AI and AI agents with Google Cloud to democratize AI into workers' hands.,Box Agent and Box AI Studio let non-technical users create custom, purpose-built AI agents over Box content.,Agents authenticate to a company's Box instance and operate within existing user permissions and security settings.,Multi-agent systems use Agent2Agent (A2A) and Box MCP server to interact with other Gemini Enterprise agents.,BigQuery is used for reporting and processing structured aspects of unstructured data.
Adya is a tech company with two primary business segments: Adya, a SaaS-based data security platform that protects data in cloud applications, and Adya.ai, an AI-powered platform that transforms enterprise workflows into AI-powered agents and copilots, improving efficiency and driving cost savings.,The company built a platform on Google Cloud that combines Gemini, Vertex AI, Google Agentspace, Google Kubernetes Engine, Pub/Sub, Cloud Storage, and Virtual Private Cloud to support secure enterprise AI workflows.,The platform includes App Studio and Model Studio for AI-assisted application generation, fine-tuned LLM development, and configurable deployment options for enterprise customers.
Enterprises have long relied on Microsoft Entra’s Conditional Access policies to safeguard organizational resources, but manually managing these policies has proven complex and error-prone due to overlapping rules, legacy authentication risks, and rapid environmental changes. The Conditional Access Optimization Agent (CAOA), powered by Security Copilot and integrated within Microsoft Entra, addresses these challenges by continuously scanning user and app inventories, sign-in logs, and existing policies to identify unprotected users, risky applications, and policy gaps. The AI-driven agent analyzes data against Zero Trust principles and provides actionable recommendations with one-click remediation, reducing manual workload while empowering administrators with full oversight. Pilot organizations have already reported freeing up security team hours and discovering over 900 unprotected users not identified through manual audits. CAOA complements existing tools—such as the What If tool and Gap Analyzer—by automating policy reviews while maintaining robust audit trails for transparency. The tool requires Microsoft Entra ID P1+ and Security Copilot licenses and currently operates in private preview. Through proactive, AI-assisted policy management, organizations can both strengthen security and reduce operational overhead.
IBM faced major productivity drag across HR, procurement, IT, supply chain, sales, and tax because employees spent time searching for policies, navigating siloed systems, handling routine support, and manually aggregating data.,IBM set out to create an enterprise-wide productivity blueprint using AI, hybrid cloud, and automation, with domain assistants such as AskIT, AskHR, and AskSales deployed across workflows.
Ericsson and AWS collaborate to deliver cloud-native, AI-powered modernized OSS/BSS for autonomous networks.,Ericsson embeds more than 20 cloud-native Telco IT AI Apps across OSS/BSS and uses Amazon SageMaker, Amazon Bedrock, Amazon Bedrock AgentCore, Amazon Q, and Kiro for generative and agentic workflows.,The article describes an Ericsson Telco Agentic AI Studio built on Amazon Bedrock, plus a Gen-AI Lab that helps fast-track ideas from experimentation to deployable products.
Tech & Comms · China
