# CrewAI Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/crewai
> Markdown URL: https://aitinkerers.org/technologies/crewai.md
> Technology record last updated: 2026-02-28T08:09:38Z
> Generated: 2026-09-21T08:39:40Z

The open-source Python framework for orchestrating autonomous, collaborative AI agent teams.

CrewAI is a lean, lightning-fast Python framework designed for building and deploying multi-agent systems. It empowers developers to define specialized Agents—each with a clear role, goal, and backstory—and organize them into a Crew to tackle complex tasks. The framework manages collaboration via context sharing and delegation, utilizing defined Processes like sequential or hierarchical execution to ensure efficient workflow. Agents gain specific capabilities through integrated features, including access to hundreds of open-source tools (e.g., internet search) and a sophisticated memory management system for short-term and long-term context. You define the agents and tasks; CrewAI handles the orchestration, delivering reliable, scalable AI automation.

- Official technology site: https://crewai.com
- Public AI Tinkerers demos and talks: 15
- Result page: 1 of 1

## Recent Public Talks and Demos

### [Evaluating Multi-Agent Systems Beyond the Final Answer](https://seattle.aitinkerers.org/talks/rsvp_lgxuusz0aeg)

I am building an evaluation framework for multi-agent AI systems that compares not only whether the system gets the right final answer, but how the agents collaborate to get there. In the live demo, I will show a small multi-agent workflow with a planner agent, worker agents, and a critic/evaluator agent, then walk through the code, traces, tool calls, handoffs, failure cases, and scoring logic used to evaluate each run. The framework captures the full agent trajectory: which agent took control, what tools were called, where context was passed or lost, how much the run cost, how long it took, and whether the final outcome matched the task goal. I will demo how the same task behaves under different setups, such as a single-agent baseline, a manager-agent architecture, and a multi-agent handoff design.

- Event context: Women in AI - Builders Track: Seattle Inaugural Meetup — 2026-07-15 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_lgxuusz0aeg

### [Multi-Agent System with Crew AI](https://amman.aitinkerers.org/talks/rsvp_PA_9D2F_iJc)

I’ll demonstrate how to build a functional multi-agent system using the CrewAI framework, showing how autonomous agents can collaborate to complete complex, multi-step tasks. The session will walk through key CrewAI concepts (agents, roles, tasks, and tools) and culminate in a live demo where multiple agents coordinate to analyze input and produce structured project outputs.

- Event context: AI Tinkerers Amman Meetup — November 15, 2025 — 2025-11-15 — Amman
- Public talk page: https://amman.aitinkerers.org/talks/rsvp_PA_9D2F_iJc

### [Building a Multi-Agent Systems for Real-World Business problems](https://nashville.aitinkerers.org/talks/rsvp_l8g7yHvZZCI)

This talk will demonstrate how to design, build, and deploy a practical multi-agent system to automate and optimize sales and marketing lead generation and evaluating user generating content. We'll move beyond theoretical concepts to address a critical business problems with actionable AI agent solutions. We'll begin by reverse-engineering the human process of identifying and qualifying leads, pinpointing the critical decision points and information gathering tasks. This analysis will guide us in defining the specific roles, responsibilities, and collaborative workflows for our AI agents. Using frameworks like crewAI and PocketFlow, we'll dive into the practical aspects of building these agents: - Defining agentic tasks and tools. - Crafting effective system and user prompts for complex, multi-step operations. - Distinguishing between simple Python/Rust automation and true agentic orchestration that adapts and strategizes. Finally, we'll cover the essential steps for real-world deployment, including containerization on AWS and robust monitoring strategies leveraging tracing tools to provide crucial operational insights for developers. Attendees will leave with a clear blueprint for applying multi-agent AI to solve tangible business challenges.

- Event context: AI Tinkerers Nashville – October 28th, 2025 — 2025-10-28 — Nashville
- Public talk page: https://nashville.aitinkerers.org/talks/rsvp_l8g7yHvZZCI

### [Writing Books with CrewAI](https://boston.aitinkerers.org/talks/rsvp_O8flPxCMcNw)

Writing Books with CrewAI explores how autonomous AI agents can transform the way we write and publish books. Instead of treating writing as a solitary word-by-word endeavor by an author, CrewAI organizes multiple agents into a coordinated "crew" that handles everything from generating outlines and drafting chapters to compiling a polished manuscript. This talk will walk through the full pipeline, idea to finished PDF, showing how to harness CrewAI’s orchestration, scripts, and binding tools to create long-form works with structure, consistency, and efficiency. You’ll see how writers can step into more of a creative-director role while AI handles the heavy lifting of research, drafting, and formatting. We'll walk through the key agents and tasks so you may understand the code flows but also how it dovetails with the author's creative and editorial process to result in a high-quality publication. CrewAI encapsulates and simply handles many details of agent orchestration and we'll look at some of the challenges faced in employing this technology and how they we're overcome. You will see research agents feed the writer agents background for their work and in real time, you'll watch the system generate full, detailed chapters. Speaker Bio Dan Hermes is the author of Prompt Power: Learn to Create ChatGPT Prompts, released last month. He works at the intersection of transformative technology and empowered teams, advancing innovation in Generative AI, Large Language Models, and software development. With a portfolio of books and articles on AI, mobile, and architecture, Dan has contributed to the growth of AI knowledge and its application across industries. Over the past decade, he has led global software and data teams at companies such as Avanade, an Accenture and Microsoft joint venture, architecting advanced applications for airlines, banks, and municipalities. For more than 20 years, he ran Lexicon Systems, a consultancy serving Microsoft, Fidelity Investments, Thermo Fisher Scientific, DraftKings, and the FAA, delivering tailored AI and software solutions. His thought leadership has appeared in IBM Mobile Business Insights and Microsoft’s MSDN Magazine, and he has spoken at countless developer user groups and venues such as IBM Think and Microsoft Ignite. Today, Dan leads projects at the forefront of GenAI, LLMs, and AI-driven applications, helping businesses harness AI’s transformative potential to build exceptional products and experiences.

- Event context: AI Tinkerers Boston: Langchain & Beyond Meetup — 2025-08-25 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_O8flPxCMcNw

### [Meta-Agents in the Wild: Building, Simulating & Scoring Generative Agents](https://medellin.aitinkerers.org/talks/rsvp_tbVIRPGQcsQ)

Lessons learned from building meta-agentic systems and using metrics-driven simulations to evaluate agents performance.

- Event context: AI Tinkerers Medellín #14 - Wizeline - 26 de Junio, 2025 — 2025-06-26 — Medellín
- Public talk page: https://medellin.aitinkerers.org/talks/rsvp_tbVIRPGQcsQ

### [Deploying AI agents without the fuss](https://miami.aitinkerers.org/talks/rsvp_bDB_iD7qv4Q)

Prototype agents are easy—running them 24/7 without blowing up your cloud bill isn’t. In this live-coding session we’ll turn a few minimal examples using Crew/Pydantic/Vercel AI SDK into a production-grade service: Pydantic for typed prompts & I/O, the Vercel AI SDK for streaming, and Agentuity’s “Heroku-for-agents” runtime for single-command deploys, autoscaling, logging, and cost controls. Expect real code, edge deploys, and a hard look at things like cross-agent comms, concurrency, and monitoring—no marketing slides, just the engineering path from laptop to live endpoint.

- Event context: AI Tinkerers - Miami: May Meetup at The LAB — 2025-05-29 — Miami
- Public talk page: https://miami.aitinkerers.org/talks/rsvp_bDB_iD7qv4Q

### [Bridging pain points to n8n workflows for non-technical users](https://seattle.aitinkerers.org/talks/rsvp_a30Qj3j3Btg)

- Demo a webapp that produces an n8n automation via conversation (AI agent chat + generative UI + artifacts) - Demo agent conversation that iteratively builds the requirements and asks user clarifying questions - Knowledge mining methods and representation for automation nodes - Anecdotes on challenges, solutions, improvements

- Event context: AI Tinkerers Seattle - April Meetup — 2025-04-25 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_a30Qj3j3Btg

### [Building an AI-Powered ML LeetCode: Multi-Agent Problem Generation & Verification](https://nairobi.aitinkerers.org/talks/rsvp_hMMSsP1KqIE)

I'll demonstrate how I built a multi-agent system that generates and verifies machine learning coding problems. The system uses multiple AI agents working together: 1. A problem generator that creates unique ML problems using GPT-4 2. A solution verifier that checks code correctness 3. A feedback generator that provides detailed improvements 4. A quality control agent that ensures problem standards

- Event context: AI Tinkerers - Nairobi Inaugural Meetup (April) — 2025-04-09 — Nairobi
- Public talk page: https://nairobi.aitinkerers.org/talks/rsvp_hMMSsP1KqIE

### [Translating Financial Data into Qualitative Insights](https://nyc.aitinkerers.org/talks/rsvp_jZXxpn6QCHo)

I will show how you can take ANY csv of financial data and use Ai to create long term structured memory using a knowledge graph that will be used to automate financial analysis. I will show the prompt structure/system and how it is used to create nodes and edges that are highly dense in context and can be retrieve to store tasks.

- Event context: AI Tinkerers NYC Meetup - March 2025 Sponsored by SignalFire and Comet — 2025-03-04 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_jZXxpn6QCHo

### [Ad video generator](https://prague.aitinkerers.org/talks/rsvp_JWnzKd4OUq0)

I will present a video generator, a project I have made on AI tinkerers hackathon in Paris. It can generate video of a real object, suitable for marketing purposes. See attached Video Demo URL.

- Event context: AI Tinkerers Prague - February 2025 Meetup — 2025-02-25 — Prague
- Public talk page: https://prague.aitinkerers.org/talks/rsvp_JWnzKd4OUq0

### [Fast, Human-In-The-Loop Agents, built via traditional APIs](https://seattle.aitinkerers.org/talks/rsvp_13lhzPPWtwQ)

We'll walk through Arch - an intelligent gateway for agents - engineered with small LLMs that offer best-in-class intent routing, guardrails, fast function calling and observability capabilities, so that developers can focus on the stuff that matters: features and business logic. Arch is built by the team that built Envoy at Lyft and offered as an open source primitive for developers.

- Event context: AI Tinkerers Seattle - December 2024 Meetup — 2024-12-12 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_13lhzPPWtwQ

### [Building Reliable Agents](https://nyc.aitinkerers.org/talks/rsvp_mnIxq4B3lhk)

How observability and monitoring enables engineers to build reliable AI agents using CrewAI

- Event context: Agent Takeover with CrewAI — 2024-07-23 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_mnIxq4B3lhk

### [LLMs and Quantitative Reasoning](https://nyc.aitinkerers.org/talks/rsvp_gtIs4hmzMk4)

LLMs cannot do quantitative reasoning, however, by using CrewAi's agent structure, this problem can be overcome by using function calling and prompt engineering. Quantitive data can be processed without the LLM ever having to process the data directly. CrewAi makes this possible through its Crews, Vector database, and LangChain.

- Event context: Agent Takeover with CrewAI — 2024-07-23 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_gtIs4hmzMk4

### [Building Autonomous AI Agents](https://toronto.aitinkerers.org/talks/rsvp_P86H2NCQSqY)

This session will feature a 5-minute live demo of an autonomous crew of AI agents, demonstrating the use of the crewAI.com open source framework, its capability to learn and make decisions in real-time. We'll also explore briefly how these agents can transform operations across industries. Demo code will be shared at https://github.com/alanwunsche/agentware-demo

- Event context: AI Tinkerers - Toronto - June 2024 Meetup — 2024-06-27 — Toronto
- Public talk page: https://toronto.aitinkerers.org/talks/rsvp_P86H2NCQSqY

### [Developing Travel Applications with Mutli-Agent Frameworks - It sounds cool but can it work?](https://seattle.aitinkerers.org/talks/rsvp_-12HEWjpF6Y)

The multi-agent toolkit space is exploding with a variety of competing offerings. Will these technologies usher in a new application development paradigm? What are the strengths and weaknesses of these nascent toolkits? Are they mature enough to support production apps? Hear from a developer who has tried several of these toolkits, building POCs in an attempt to understand if they are mature enough to support real-world travel applications.

- Event context: AI Tinkerers Seattle - April 2024 Meetup — 2024-04-26 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_-12HEWjpF6Y

## Related Technologies

- [LangChain](https://aitinkerers.org/technologies/langchain) ([Markdown](https://aitinkerers.org/technologies/langchain.md)) — 445 public demos
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [LangGraph](https://aitinkerers.org/technologies/langgraph) ([Markdown](https://aitinkerers.org/technologies/langgraph.md)) — 67 public demos
- [Agentuity](https://aitinkerers.org/technologies/agentuity) ([Markdown](https://aitinkerers.org/technologies/agentuity.md)) — 1 public demo
- [agentware-demo](https://aitinkerers.org/technologies/agentware-demo) ([Markdown](https://aitinkerers.org/technologies/agentware-demo.md)) — 1 public demo
- [Anthropic API](https://aitinkerers.org/technologies/anthropic-api) ([Markdown](https://aitinkerers.org/technologies/anthropic-api.md)) — 66 public demos
- [Apache Airflow](https://aitinkerers.org/technologies/apache-airflow) ([Markdown](https://aitinkerers.org/technologies/apache-airflow.md)) — 2 public demos
- [Arch](https://aitinkerers.org/technologies/arch) ([Markdown](https://aitinkerers.org/technologies/arch.md)) — 1 public demo
- [archgw](https://aitinkerers.org/technologies/archgw) ([Markdown](https://aitinkerers.org/technologies/archgw.md)) — 1 public demo
- [Autogen](https://aitinkerers.org/technologies/autogen) ([Markdown](https://aitinkerers.org/technologies/autogen.md)) — 6 public demos
- [AWS Lambda](https://aitinkerers.org/technologies/aws-lambda) ([Markdown](https://aitinkerers.org/technologies/aws-lambda.md)) — 14 public demos
- [Browserbase](https://aitinkerers.org/technologies/browserbase) ([Markdown](https://aitinkerers.org/technologies/browserbase.md)) — 3 public demos
- [CopilotKit](https://aitinkerers.org/technologies/copilotkit) ([Markdown](https://aitinkerers.org/technologies/copilotkit.md)) — 8 public demos
- [Crews](https://aitinkerers.org/technologies/crews) ([Markdown](https://aitinkerers.org/technologies/crews.md)) — 1 public demo
- [Envoy](https://aitinkerers.org/technologies/envoy) ([Markdown](https://aitinkerers.org/technologies/envoy.md)) — 1 public demo
- [Flux](https://aitinkerers.org/technologies/flux) ([Markdown](https://aitinkerers.org/technologies/flux.md)) — 8 public demos
- [function calling](https://aitinkerers.org/technologies/function-calling) ([Markdown](https://aitinkerers.org/technologies/function-calling.md)) — 11 public demos