Google Cloud Opens 'Gemini Factory' to Speed Partner AI Agent Builds
Google Cloud's Gemini Factory puts its engineers beside software partners in seven hubs to build and list Gemini agents, connectors, skills and plugins on Google Cloud Marketplace.

Why it matters
- Google Cloud launched the Gemini Factory partner programme on 18 November 2026 to help vendors build and list Gemini-based integrations on Google Cloud Marketplace.
- Hands-on workshops run in seven hubs: San Francisco, New York City, Sunnyvale, London, Paris, Tel Aviv and Tokyo.
- Partners can use four building blocks: MCP connectors, Markdown-based declarative skills, plugins and standalone AI agents in partner-controlled environments.
- Nine early partners including Aily Labs, FlowX, Klarvant, NextLM, Optimizely, Ping Identity, Sinch, THG Ingenuity and Xiatech have built production integrations.
- Partners receive seed credits to offset build costs, plus performance credits and marketing support for qualified listings.
Google Cloud has launched the Gemini Factory, a partner programme that puts Google engineers side by side with enterprise software vendors to build, test and list Gemini-based AI agents, connectors, skills and plugins on Google Cloud Marketplace. The announcement, made on 18 November 2026, signals how sharply the cloud division is courting third-party developers as distribution becomes the deciding battleground among AI platform providers.
The stakes are straightforward. Enterprise customers increasingly buy AI capability through marketplaces rather than bespoke projects, and vendors that can turn a use case into a listed integration fastest capture that procurement channel. Google Cloud is betting that hands-on engineering support, plus financial incentives, will pull partners onto Gemini ahead of rival agent platforms.
What does the programme actually provide?
The Gemini Factory centres on in-person workshops in seven technology hubs: San Francisco, New York City, Sunnyvale, London, Paris, Tel Aviv and Tokyo. Google Cloud specialists work directly with partner teams to design, build and prepare integrations for distribution to enterprise customers.
The programme targets a specific pain point: the engineering work needed to turn a use case into a shippable product. Google Cloud said that work includes selecting an architecture, setting up authentication, managing multi-step tool execution and creating in-line user experiences inside Gemini. Each of those steps has historically slowed enterprise AI pilots from reaching production.
Which building blocks can partners use?
Google Cloud has outlined four main integration paths:
- Model Context Protocol connectors that expose external APIs and data tools to Gemini.
- Declarative skills written in Markdown, designed for repeatable workflows and guardrails.
- Plugins that package connectors and skills together for governance and use.
- Standalone AI agents that run in a partner-controlled environment while connecting back into Gemini.
The standalone-agent option carries particular weight for regulated industries. Partners with proprietary algorithms or regulated processes can keep their agents in isolated environments while still calling them through Gemini. Google Cloud said the design protects intellectual property and maintains audit trails — two requirements that have kept many financial services and healthcare vendors from embedding their logic inside third-party AI runtimes.
The development path also includes the Antigravity environment and what Google Cloud describes as rapid agent development workflows. Partners use those tools to create remote connectors, develop skills, test integrations and prepare them for ingestion into the wider Gemini environment.
How do partners get paid and distributed?
The commercial route runs through Google Cloud Marketplace. Beyond product development, the programme includes support for marketplace publication and sales activity. Partners can receive seed credits to offset build and porting costs, followed by performance credits and marketing support for qualified listings, Google Cloud said.
Distribution does not stop at the marketplace itself. Partner tools also appear through Google Cloud's AI Solution Finder, which gives administrators a central point for configuration and governance — a nod to buyers who need policy control over which agents employees can invoke.
Customer onboarding works through native OAuth, Dynamic Client Registration and bring-your-own-licence flows. Under that model, partners keep control of backend security, data sources and proprietary tooling in their own environments, rather than handing them to a platform runtime.
Who shipped first?
Google Cloud highlighted a group of early vendors that have already built production integrations through the programme, spanning decision intelligence, financial services, cyber governance, sales prospecting, marketing operations, identity management, messaging, retail analytics and order processing:
- Aily Labs built a decision intelligence agent for senior executives that draws on operational and financial data to forecast trends and identify disruptions.
- FlowX developed a capital markets document reconciliation agent focused on document extraction, reconciliation and loan-pack auditing.
- Klarvant created a governance analyst agent that examines an enterprise's external namespace for issues such as certificate hygiene, email security and post-quantum readiness.
- NextLM introduced a prospecting agent that ranks business accounts and buyers using behavioural data across hundreds of millions of companies and professionals.
- Optimizely contributed a marketing calendar deck builder.
- Ping Identity built a device management agent for employee self-service around MFA devices.
- Sinch shipped an RCS messaging agent.
- THG Ingenuity produced a shopping insights tool.
- Xiatech delivered an order intake agent for commerce and supply-chain workflows.
Several partners said the programme substantially shortened build times, and three offered detailed accounts of why.
What did the early partners say?
Reynaldo Espana, VP of Engineering at Aily Labs, described the workshop as a turning point for the company's flagship agent. "Our mission is to bring real-time, AI-powered decision intelligence to every business executive. The Agent Factory workshop was an incredible catalyst; it enabled our team to wire up complex predictive reasoning models into the Gemini Enterprise runtime, activating Aily Super Agent in record time, orchestrating thousands of specialized agents, connecting insights across multiple business domains, and presenting them through a single, intuitive interface for the user. Having side-by-side engineering support helped us prove out high-impact executive scenarios and accelerate our roadmap toward a verified listing on Google Cloud Marketplace," Espana said.
FlowX, working in capital markets, stressed compliance as much as speed. "In capital markets and lending, precision, auditability and data security are non-negotiable. What impressed us most about the Agent Factory was how rapidly we could implement complex business validation rules and multi-source document ingestion without compromising on enterprise compliance. Transitioning from raw use cases to a functional prototype in a single workshop demonstrated the sheer execution velocity possible on Gemini," said Serban Chiricescu, Founder and Chief Technology Officer of FlowX.AI. A prototype in one workshop sets the benchmark the programme will now be measured against.
Ping Identity framed its build around the identity events that flood corporate help desks. "Every new phone and every last day at work is an identity event, and too many of them still turn into help desk tickets. Working side by side with Google Cloud in the Agent Factory, we built agents that let employees manage their own MFA devices and let admins resolve identity requests in one conversation. With ADK, A2A and A2UI, those experiences run natively in Gemini, and every action stays scoped to the user, confirmed by a person, and accountable," said Loren Russon, SVP, Product and Technology at Ping Identity. The mention of ADK, A2A and A2UI points to the protocol stack — Agent Development Kit, agent-to-agent and agent-to-UI — that Google Cloud expects partners to build against.
Why the marketplace push matters now
The programme lands as enterprise AI adoption shifts from pilots to procurement. Marketplaces give cloud vendors a compounding advantage: every listed integration makes the underlying platform stickier for the customer and more attractive to the next vendor. The credits, marketing support and AI Solution Finder placement are the levers Google Cloud is using to stock that shelf.
The early partner list suggests the strategy is working in regulated and data-heavy sectors, where integration complexity has been the highest barrier. Whether the Gemini Factory can sustain that velocity across a broader vendor base will determine whether Gemini becomes the default runtime for enterprise agents or one option among several.
Source: itbrief.com.au
More from Rebecca Stone
Show full bio
Correspondent covering companies, deals and strategy in artificial intelligence and machine learning at AI In Context.
728 articles
Related articles
- Google turns Gemini into an agent that works like a co-worker
- Google Makes Tasks in Gemini Free Across Six Countries
- Google’s Gemini for healthcare targets the paperwork behind care
- Google Opens Gemini 4 Argon to All Paying Users at $2 Per Million Tokens
- Google ships Gemini 3 Pro with a 1501 Elo LMArena score