Cloud-based Software Development Kits

Google just released a full-stack toolkit for building AI agents.. and it’s a big deal. 🚀 Until now, building production-grade agents has felt like duct-taping together libraries: One for logic, another for tools, and almost nothing for evaluation or deployment. That changes with Google’s new open-source Agent Development Kit (ADK), an end-to-end operating system for building, testing, and shipping intelligent agents. Here’s why this release stands out: 🔧 Code-first, developer-focused Built for serious devs who need version control, custom logic, and robust testing. 🤖 Multi-agent, by design Easily spin up systems where agents collaborate or specialize across tasks—right out of the box. 🧪 Goes beyond building Most frameworks stop at the prototype. ADK includes tools for evaluating performance and deploying workflows into production. 🧩 Flexible orchestration Define custom flows using built-in agents, or wire up your own with dynamic routing logic. 💻 Great local dev experience CLI + Web UI make it easy to build, test, and debug your agents locally—before pushing to prod. Bonus: It’s cloud-friendly (of course it works well with Google Cloud), but supports any third-party models and tools, so you’re not locked in. To get started: pip install google-adk GitHub repo is linked in the comments👇

Google Agent Development Kit (ADK) - yet another agent framework (YAAF ��) ? I know what you're thinking - does the world really need another agent framework? Well, let me try to make a case for it 🙃 First off, please check out the demo below from Franziska Hinkelmann, Ph.D. from Google's Cloud Next Developer Keynote. She builds a pretty impressive proposal agent live on stage, and in doing so, really shows off some of ADK's highlight features: ⚙️ 𝙎𝙩𝙖𝙣𝙙𝙖𝙧𝙙𝙞𝙨𝙚𝙙 𝙏𝙤𝙤𝙡 𝙄𝙣𝙩𝙚𝙜𝙧𝙖𝙩𝙞𝙤𝙣 In the demo, Fran defines an agent using LLMAgent, letting Gemini handle the reasoning. But ADK doesn't lock you into just LLM-driven decisions. It also provides WorkflowAgents (for strict sequences, loops, parallel tasks) and CustomAgents where you write the exact Python orchestration logic. This is key because it lets you blend deterministic control (vital for reliable business processes) with AI flexibility, offering a level of precision that's sometimes harder to guarantee when only prompting an LLM for complex flows. 🔌 𝙏𝙤𝙤𝙡 𝙪𝙨𝙚 In the demo Fran highlights the ADK's built-in support for Anthropic's Model Context Protocol (MCP). The agent uses MCP to perform RAG against a private database of building codes via Google's open-source MCP Toolbox for Databases. This shows ADK integrating with standardised protocols for robust connections to our own data and APIs, alongside its built-in tools (like Google Search, Code Execution) and wrappers for tools from LangChain/CrewAI. ✨ 𝘿𝙚𝙫𝙚𝙡𝙤𝙥𝙚𝙧 𝙀𝙭𝙥𝙚𝙧𝙞𝙚𝙣𝙘𝙚 (𝘿𝙓) ADK comes with a neat local Dev UI which is more than just a chatbot – it can be used for testing, inspecting events, and handles multimodal input by uploading both text and a floor plan image. Behind the scenes, ADK provides integrated concepts like SessionService (handling conversation state/history) and ArtifactService (managing files, like the PDF the agent generated), giving you essential application plumbing out-of-the-box. ☁️ 𝘿𝙚𝙥𝙡𝙤𝙮𝙢𝙚𝙣𝙩 Fran's initial diagram shows ADK within Vertex AI Agent Engine. ADK is designed with production on Google Cloud in mind, offering easy deployment paths to managed services like Agent Engine or Cloud Run. But critically, it also provides the flexibility for standard Docker/Kubernetes support if you prefer self-deployment. If you're building complex agents where precision, integration, and a code-first approach matter, ADK offers some compelling advantages worth exploring.

Building AI agents is one thing. Getting them into production? That’s when agent engineering demands enterprise architecture. You shouldn't have to rewrite your entire stack just to host a sophisticated multi-agent system. The Agent Development Kit (ADK) is completely open-source, natively supporting Python, Go, TypeScript, Java, and Kotlin. No matter what language your team writes in, Google Cloud's Agent Platform meets you where your team already works. It gives you a complete spectrum of options, from a fully managed runtime to maximum infrastructure control. Here is how you can ship. 1️⃣ The Fast Track Building in Python or Go? Use native deployment tools `adk deploy` or `adkgo deploy`. Push your code straight to Agent Runtime. It is a fully managed, serverless environment optimized specifically for agents. It abstracts the operational complexities away. You get integrated sessions, memory, sandboxing, and strict enterprise security instantly. 2️⃣ The Flexible Path Need to deploy TypeScript, Java, Python, Go, or any languages or highly customized workflows? Bring Your Own Container (BYOC) is your universal strategy. Ship your custom Dockerfile via Artifact Registry directly to Agent Runtime to keep all those integrated cognitive services. Or, deploy that same container to Cloud Run for a purely serverless, stateless execution model. With Cloud Run, you interaction with the agent is easily facilitated over standard, universally compatible HTTP endpoints. 3️⃣ The Heavy Duty Building complex agent meshes? Need hardware accelerators? Deploy straight to Google Kubernetes Engine (GKE) for maximum control. GKE enables the hosting of localized, open-weight models directly adjacent to your agent logic. This option is perfectly suited for mature DevOps organizations that need to operate multi-agent systems at an immense scale. 4️⃣ The Sovereign Route Working in highly regulated environments? Package your agents into generic Docker or Podman containers. Run them absolutely anywhere. It is the perfect setup for offline, air-gapped systems that require strict data sovereignty outside of external clouds. No vendor lock-in. Just pure, developer-first infrastructure. Write the code. Pick your environment. Ship it. Enterprise architecture shouldn't be a roadblock to your agent engineering. It should be the launchpad. ✦ ADK Deploy: https://adk.dev/deploy ✦ Agent Runtime: https://lnkd.in/gS6iqx9e ✦ Cloud Run: https://lnkd.in/gsxWDe4U ✦ GKE: https://lnkd.in/g3rMfbRa

𝐖𝐞 𝐡𝐚𝐯𝐞 𝐚𝐥𝐥 𝐬𝐞𝐞𝐧 𝐭𝐡𝐨𝐬𝐞 𝐠𝐢𝐚𝐧𝐭 𝐘𝐀𝐌𝐋 𝐟𝐢𝐥𝐞𝐬, 𝐏𝐚𝐠𝐞𝐬 𝐚𝐧𝐝 𝐩𝐚𝐠𝐞𝐬 𝐨𝐟 𝐜𝐨𝐧𝐟𝐢𝐠𝐮𝐫𝐚𝐭𝐢𝐨𝐧. You change one line, and… boom. Chaos. But what if I told you, you could program your cloud infrastructure the same way you write your app code? That is exactly what AWS CDK does. 𝐋𝐞𝐭’𝐬 𝐛𝐫𝐞𝐚𝐤 𝐢𝐭 𝐝𝐨𝐰𝐧: • AWS CDK (Cloud Development Kit) lets you define your cloud setup using real code - TypeScript, Python, Java. • No more endless JSON or YAML. You use loops, conditions, functions. It’s just code. 𝐖𝐡𝐲 𝐞𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐬 𝐚𝐫𝐞 𝐥𝐨𝐯𝐢𝐧𝐠 𝐭𝐡𝐢𝐬: ✅ 𝐂𝐨𝐝𝐞-𝐧𝐚𝐭𝐢𝐯𝐞 𝐞𝐱𝐩𝐞𝐫𝐢𝐞𝐧𝐜𝐞 You can reuse logic, create abstractions, and think like a developer. ✅ 𝐑𝐞𝐮𝐬𝐚𝐛𝐥𝐞 𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐬 Build shareable “𝐢𝐧𝐟𝐫𝐚 𝐜𝐨𝐦𝐩𝐨𝐧𝐞𝐧𝐭𝐬” - like reusable Lego blocks for your cloud. ✅ 𝐏𝐫𝐞𝐯𝐢𝐞𝐰 𝐛𝐞𝐟𝐨𝐫𝐞 𝐝𝐞𝐩𝐥𝐨𝐲𝐢𝐧𝐠 Use`𝐜𝐝𝐤 𝐝𝐢𝐟𝐟`to see exactly what will change before you hit deploy. ✅ 𝐌𝐮𝐥𝐭𝐢-𝐬𝐭𝐚𝐜𝐤, 𝐦𝐮𝐥𝐭𝐢-𝐞𝐧𝐯 𝐫𝐞𝐚𝐝𝐲 Handle complex production, staging, and dev environments without headaches. ✅ 𝐒𝐭𝐫𝐨𝐧𝐠 𝐭𝐲𝐩𝐢𝐧𝐠 & 𝐬𝐚𝐟𝐞𝐭𝐲 Catch errors before runtime with compile-time checks. 𝐖𝐡𝐚𝐭 𝐜𝐚𝐧 𝐲𝐨𝐮 𝐛𝐮𝐢𝐥𝐝? • Full serverless apps (Lambda, API Gateway, DynamoDB) • VPCs, ECS clusters, ALBs • Event-driven pipelines • CI/CD systems 𝐓𝐡𝐞 𝐛𝐢𝐠 𝐬𝐡𝐢𝐟𝐭? You are no longer just describing your infrastructure. You are programming it. And compared to Terraform or plain CloudFormation, CDK feels like moving from a flip phone to a smartphone. --- 💬 𝐇𝐚𝐯𝐞 𝐲𝐨𝐮 𝐭𝐫𝐢𝐞𝐝 𝐂𝐃𝐊? What is been your biggest “𝐚𝐡𝐚” moment (or biggest headache)? Drop your thoughts below. Let’s unpack it together.

Google’s Agent Development Kit (ADK) - an open-source, flexible framework designed to simplify the creation and deployment of AI agents and multi-agent systems. With ADK, developers can design intelligent agents that think, act, and coordinate, whether it’s for conversational assistants, automation, or complex multi-step workflows. What sets ADK apart is its model-agnostic and deployment-agnostic design. Though optimized for Google’s Gemini and cloud stack, it supports other LLMs, tools, and infrastructures as well. It makes agent development feel more like building software - structured, modular, and highly adaptable. Core Concepts An Agent in ADK is a self-contained unit capable of reasoning, using tools, and collaborating with other agents. Developers can create LLM-based agents for natural language tasks, workflow agents for automation, or custom agents for domain-specific logic. Tools & Ecosystem ADK supports a rich tool ecosystem, including pre-built utilities for search, code execution, and custom functions. Agents can even use other agents as tools, allowing highly modular and scalable AI architectures. Orchestration & Workflow Developers can control agent behavior using workflow types like Sequential, Parallel, or Loop, or rely on LLM-driven routing for dynamic orchestration. This combination enables hybrid systems that balance rule-based logic with adaptive intelligence. Deployment Options Once built, agents can be packaged into containers and deployed across environments, from Vertex AI Agent Engine and Cloud Run to custom infrastructures like Docker, GKE, or on-prem servers. Ready to experiment with AI agents? Install it with: pip install google-adk Then build your first multi-agent application using Python or Java and deploy it wherever you want. #AgentDevelopmentKit