AI development continues to change with the consistent release of new models, standards, and system architectures. It can often be a lot to keep track of and learn. But DigitalOcean has you covered with our community tutorials and resources.
These 10 tutorials from last month cover both practical, hands-on topics (such as building a game with GPT-5.4) and explanatory concepts (like migrating to multi-agent systems). Take a look and try them out—or bookmark them for some weekend coding!
Getting Started with Qwen3.5 Vision-Language Models
This tutorial walks through how to run and experiment with Qwen 3.5, an open-source multimodal model family that handles text, images, and even video. It breaks down the model’s architecture and demonstrates how to deploy it on GPU infrastructure so you can build apps like coding assistants or document analyzers on your own stack. You’ll see how high-performing multimodal AI is becoming accessible without relying on proprietary APIs.
A2A vs MCP: How These AI Agent Protocols Actually Differ
Read about the difference between two emerging standards for agent-based systems: agent-to-agent communication (A2A) and model context protocol (MCP). You’ll learn when to use each—A2A for coordinating multiple agents and MCP for structured tool integration—and why most production systems combine both. It’s a practical breakdown of the protocols shaping how agentic AI systems are actually built.
Nemotron 3 Helped Me Find the Perfect Dish Rack?
Get insight into how NVIDIA’s Nemotron 3 model pairs with NemoFinder to improve retrieval and reasoning workflows. This tutorial demonstrates how combining LLMs with optimized search and ranking pipelines can yield more accurate results, especially in enterprise or knowledge-intensive applications. You’ll also learn more about how retrieval-augmented generation (RAG) systems are evolving with tighter model–tool integration.
Train YOLO26 for Retail Object Detection on DigitalOcean GPUs
This hands-on guide shows how to train a YOLOv26 model for retail use cases such as shelf monitoring and product detection on GPU infrastructure. It walks through dataset prep, training, and deployment so you can build real-world computer vision pipelines. You’ll gain a better understanding of how to move from raw image data to a production-ready detection model.
Building Long-Term Memory in AI Agents with LangGraph and Mem0
If you’re curious about how to add persistent memory to agent workflows using LangGraph and Mem0, check out this tutorial. It shows how agents can retain context across sessions, enabling more personalized and stateful interactions over time. Its key takeaway is how long-term memory transforms agents from stateless responders into systems that can learn and adapt.
Crafting a Game from Scratch with GPT-5.4
This article breaks down GPT-5.4’s capabilities, improvements, and practical use cases. It highlights advancements in reasoning, efficiency, and multimodal performance, and shows how developers can integrate the model into real applications. You’ll see how this frontier model integrates into modern AI stacks and the steps involved in creating a 3D badminton game from the ground up.
What are Text Diffusion Models? An Overview
This guide introduces diffusion models for text generation and explains how they differ from traditional autoregressive LLMs. It walks through how diffusion-based approaches iteratively refine outputs and where they may outperform standard models. You’ll get a conceptual and practical understanding of an emerging alternative to transformers.
LLM Tool Calling with Gradient™ AI Platform and Databases
Discover how to connect LLMs to external tools—like databases—using structured tool calling. It walks through building workflows in which models query, retrieve, and act on real data rather than relying solely on prompts. You’ll get to see that tool integration makes LLMs more reliable and production-ready.
How to Generate Videos with LTX-2.3 on DigitalOcean GPU Droplets
This tutorial explores how to generate videos using LTX 2.3, covering setup, prompts, and rendering workflows. It demonstrates how generative AI is expanding beyond text and images into video creation. After this article, you’ll know how to experiment with video generation pipelines and integrate them into creative or product workflows.
From Single to Multi-Agent Systems: Key Infrastructure Needs
Get an overview of what changes when you move from a single AI agent to a multi-agent system. This tutorial goes through the full infrastructure stack—covering orchestration patterns, communication protocols, memory, and observability—so you can design systems where multiple agents collaborate reliably. Ultimately, multi-agent setups unlock scalability and specialization but require significantly more coordination, state management, and fault tolerance to work in production.
Top comments (0)