> Markdown version of [/videos/1725-wearedevelopers-live-build-a-multi-ai-agents-game-master-with-strands-our-weekly-web-finds?t=2031](https://www.wearedevelopers.com/videos/1725-wearedevelopers-live-build-a-multi-ai-agents-game-master-with-strands-our-weekly-web-finds?t=2031). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# WeAreDevelopers LIVE - Build a multi AI agents game master with Strands & our weekly web finds

Tiffany Soter proves true AI orchestration replaces monolithic prompts. Watch her build a dynamic, multi-agent D&D Game Master using the Strands SDK. Give your ChatGPT hands today.

- **Speakers:** [Chris Heilmann](https://www.wearedevelopers.com/@chris-heilmann), [Daniel Cranney](https://www.wearedevelopers.com/@daniel-cranney), Tiffany Souterre
- **Event:** WeAreDevelopers LIVE
- **Published:** October 1, 2025
- **Duration:** 1:32:50
- **URL:** https://www.wearedevelopers.com/videos/1725-wearedevelopers-live-build-a-multi-ai-agents-game-master-with-strands-our-weekly-web-finds

## Summary

The discussion kicks off by analyzing the latest developer trends, challenging the viability of mobile "vibe coding" environments compared to robust desktop setups, and evaluating Microsoft's new AI marketplace ambitions. Emphasizing the persistent need for developer due diligence, the hosts discuss recent NPM supply chain vulnerabilities and the necessity of understanding underlying code, even as automation capabilities increase. The conversation also touches on the cognitive benefits of "cozy gaming" and digital detoxes, suggesting that genuine human creativity requires spaces completely devoid of algorithmic doom-scrolling.

The core narrative centers on AWS Senior Developer Advocate Tiffany Soter, who demonstrates how to construct a dynamic, multi-agent Dungeons and Dragons Game Master using the newly open-sourced Strands SDK. Tiffany expertly illustrates the evolution from conversational LLMs to action-oriented agents, characterizing an agent simply as "a ChatGPT with hands." The demo walks through connecting an agent to executable actions via the Model Context Protocol (MCP), essentially establishing a standardized API design tailored specifically for agents. By spinning up an MCP server for a custom D20 dice-rolling function alongside a specialized character-creation agent on an A2A (Agent-to-Agent) server, the walkthrough highlights how modern AI can dynamically discover tools and delegate complex generative logic.

Despite the inherent chaos of live coding without autocomplete, the walkthrough reveals a profound shift in software architecture. Rather than relying on monolithic system prompts, developers can now build orchestrators that seamlessly retrieve external database context, execute scripts, and organically collaborate with specialized AI expert nodes. Tiffany predicts an upcoming cultural shift from traditional application platforms to an "agent store" ecosystem, where secure, modular assistants natively cooperate to function as an "actual Jarvis." This practical methodology proves that scalable, cross-collaborative AI is not merely a novelty for tabletop gaming, but represents the foundational architecture for the next era of autonomous application development.

**Keywords:** strands sdk, model context protocol, mcp server implementation, a2a communication protocol, multi-agent systems, ai orchestrator patterns, aws bedrock integration, local llm deployment, ai agent tool creation, npm supply chain vulnerabilities, vibe coding challenges, agent store ecosystem, automated autonomous workflows, digital detox for developers, dungeons and dragons ai

## Chapters

1. **Submitting proposals for upcoming developer conferences** (00:07) — Guidelines for submitting high-quality event proposals for developer congresses located in major technological hubs.
1. **Using large language models as tabletop game masters** (04:24) — How artificial intelligence supports role-playing campaigns by dynamically generating characters and managing unpredictable player choices.
1. **Astrobiology and finding evidence of life on Mars** (07:28) — Defining characteristics of microbacterial life inside soil samples mimicking the severe environmental conditions on planetary neighbors.
1. **The limitations of writing code on mobile devices** (11:10) — Why vibe coding applications natively run on smartphones struggle against traditional desktop development setups.
1. **Evaluating the microsoft marketplace for publisher extensions** (15:40) — Obstacles around independent monetization stores integrating directly into established corporate artificial intelligence suites.
1. **Addressing supply chain compromises in open source registries** (19:32) — Efforts by infrastructure providers to intercept malicious packages infiltrating widely used dependency registries.
1. **Corporate influence and governance in open source communities** (23:15) — The impact of corporate takeovers replacing foundational library maintainers throughout collaborative development spaces.
1. **Retraining workforce staff for the artificial intelligence era** (25:07) — Navigating business decisions around upskilling employees to utilize generative tools naturally within daily workflows.
1. **Visualizing complex algorithms as furniture assembly instructions** (28:29) — Human-generated visual guides creatively unpacking abstract programming concepts through textless vector drawings.
1. **Evaluating user interfaces through impaired interaction states** (30:20) — Simulating impaired vision patterns reveals potential accessibility barriers and hidden navigational problems affecting web implementations.
1. **Selecting optimal typography for code editors** (32:19) — Utilizing tournament-style elimination tools to determine readable script formatting eliminating intricate styling.
1. **Restoring cognitive focus through unstructured digital experiences** (33:51) — Substituting high-paced gameplay interactions with open-ended digital systems effectively combats systemic computational burnout.
1. **Generating personal business cards in the command line** (38:21) — Packaging graphical developer portfolios deploying standardized commands resulting into rendered terminal summaries.
1. **Comparing token pricing across language model providers** (39:45) — Evaluating computational expenditures and hidden provider costs when utilizing conversational string generators.
1. **Exploring model context protocols for frontend web resources** (40:50) — Aggregating community standards around interface rules designed specifically for optimizing connections inside web platforms.
1. **Optimizing array searches with native javascript utilities** (41:41) — Replacing repetitive manual inversions implementing novel traversal iterations heavily improving total collection execution timings.
1. **Linking application deletion instances to operating system bins** (42:35) — Applying integrated reversal options redirecting application deletions immediately toward safe file recovery buckets safely.
1. **Understanding language models and autonomous executing agents** (44:27) — Extending conversational boundaries transforming prompt processors fully toward systems utilizing active technical integrations securely.
1. **Standardizing tool connections utilizing model context protocol** (47:11) — Abstracting complex logic layers standardizing universal interfaces communicating dynamically across external model endpoints.
1. **Defining standards for agent to agent communication** (48:55) — Establishing interaction pipelines instructing independently operated software instances handling separated objectives efficiently.
1. **Initializing basic language models using the strands framework** (50:13) — Writing basic python definitions booting provider-agnostic engines enabling seamless switches over different background structures.
1. **Configuring execution environments utilizing locally hosted models** (53:24) — Binding explicit web endpoints linking internally hosted processors completely replacing traditional external web computations.
1. **Providing executable tooling endpoints to existing running models** (55:18) — Mapping community utilities allows logical processors independently resolving constraints required constructing custom computations dynamically.
1. **Designing and embedding customized operational functions for agents** (60:52) — Generating strict parameter controls wrapping traditional application scripts granting programmatic control exposing new properties.
1. **Exposing operational capabilities through dedicated protocol servers** (64:41) — Launching secondary internal connections authorizing separate operational instances securely accessing localized function rules continually.
1. **Connecting language models natively into remote operational servers** (67:20) — Fetching available definitions synchronously parsing dynamically served instructions providing expanded functional operations across scripts.
1. **Structuring agent collaborations leveraging customized server interfaces** (70:01) — Instantiating sequenced operations managing structured data distributions directing separated tasks between isolated processing bots.
1. **Demonstrating orchestrated actions natively resolving inside graphical views** (81:54) — Presenting customized consumer solutions effectively interpreting sequential commands evaluating background actions directly within applications.
1. **Reviewing future paradigm shifts regarding autonomous digital workflows** (85:10) — Predicting emerging digital platforms transitioning towards specialized automated systems autonomously communicating complex daily objectives.

## Related Moments

- [Rethinking team structures around AI agent capabilities](https://www.wearedevelopers.com/videos/1539-agentic-devops-how-ai-powered-automation-transforms-software-delivery-on-github-and-azure) (from "Agentic DevOps: How AI-Powered Automation Transforms Software Delivery on GitHub and Azure")
- [Moving from playing with artificial intelligence to business implementation](https://www.wearedevelopers.com/videos/2136-moving-from-playing-with-ai-to-implementing-ai-lee-stott) (from "Moving from Playing with AI to Implementing AI - Lee Stott")
- [Shifting developer workloads and realistic AI productivity gains](https://www.wearedevelopers.com/videos/1830-wearedevelopers-live-speculaitions) (from "WeAreDevelopers LIVE - SpeculAItions")
- [The evolving role of software engineers alongside agents](https://www.wearedevelopers.com/videos/100132-the-agent-interface-layer-protocols-tools-and-trust-boundaries) (from "The Agent Interface Layer: Protocols, Tools and Trust Boundaries")
- [Discussion on AI hallucinations and practical developer workflows](https://www.wearedevelopers.com/videos/805-openai-for-fintech-building-a-stock-market-advisor-chatbot) (from "OpenAI for FinTech: Building a Stock Market Advisor Chatbot")
- [Balancing developer autonomy with the adoption of coding agents](https://www.wearedevelopers.com/videos/100198-the-last-mile-of-ai-from-prototype-to-production) (from "The Last Mile of AI: From Prototype to Production")

## Related Articles

- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)
- [Dev Digest 132 - Binging WADFlix?](https://www.wearedevelopers.com/magazine/473-dev-digest-132-binging-wadflix)
- [Exploring AI: Opportunities and Risks for Developers](https://www.wearedevelopers.com/magazine/522-exploring-ai-opportunities-and-risks-for-developers)
- [Dev Digest 137 - AI'm not sure about this](https://www.wearedevelopers.com/magazine/485-dev-digest-137-ai-m-not-sure-about-this)

## Related Jobs

- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Principal Product Manager, Agent Platform](https://www.wearedevelopers.com/jobs/ext/277541-principal-product-manager-agent-platform) at **GitHub**
- [Principal Field Architect - AI Agents](https://www.wearedevelopers.com/jobs/ext/1442858-principal-field-architect-ai-agents) at **Twilio**
- [Senior AI Agent Software Engineer (Go, Python) (m/f/x)](https://www.wearedevelopers.com/jobs/48277-senior-ai-agent-software-engineer-go-python-m-f-x) at **Dynatrace**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**