# AI agents Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/ai-agents
> Markdown URL: https://aitinkerers.org/technologies/ai-agents.md
> Technology record last updated: 2026-02-25T07:26:53Z
> Generated: 2026-09-22T07:46:12Z

Autonomous software systems that leverage LLMs to reason, plan, and execute complex, multi-step goals across external tools and data sources.

AI agents are the next evolution of applied AI: they are goal-driven, autonomous entities that handle entire workflows without constant human prompting. Unlike simple chatbots, agents use a core LLM for reasoning, integrate with external tools (APIs, CRMs, web browsers), and maintain memory to adapt and improve over time. Enterprises deploy them for high-value automation: examples include a sales agent generating over 2,000 qualified leads monthly or a research agent analyzing 50 petabytes of clinical data for insights. This technology is about scaling complex decision-making and action, not just conversation.

- Official technology site: https://cloud.google.com/vertex-ai
- Public AI Tinkerers demos and talks: 35
- Result page: 1 of 2

## Recent Public Talks and Demos

### [Manifest: An open-source LLM router that focuses on reducing AI inference costs](https://geneva.aitinkerers.org/talks/rsvp_1OM9lldCTbk)

Manifest is a smart model router for agents and AI apps that redirects each query to the right model, saving up to 70% in AI costs. 🔀 Routing based on complexity, specificity and custom HTTP headers 🎛️ Mix your providers: API keys, Subscriptions, Local models, Custom providers 📊 Track every single dollar, setup notifications and limits 🚑 Fallback on different models when queries fails

- Event context: AI Tinkerers Geneva Inaugural Meetup - May 2026 — 2026-05-13 — Geneva
- Public talk page: https://geneva.aitinkerers.org/talks/rsvp_1OM9lldCTbk

### [Compose and Dragons: Tiny Language Models in Action](https://paris.aitinkerers.org/talks/rsvp_lNiq-CojExE)

Let's debunk some beliefs about (very) small LLMs, those that make less than 4b of parameters. We often hear: They are useless and do not know how to do anything, they know nothing, they are bad at calling (so no MCP) This is partly wrong, and we can fix the rest and build generative AI systems with these very small models. Among other things, we will see how to create NPCs with a personality, a master dungeon that will manage your movements, fights... and allow you to talk to this or that NPC ...

- Event context: AI Tinkerers Paris: Docker Agentic Workflows (Devoxx Kickoff) — 2026-04-21 — Paris
- Public talk page: https://paris.aitinkerers.org/talks/rsvp_lNiq-CojExE

### [Anthropic Ambassador Chile](https://santiago.aitinkerers.org/talks/rsvp_HCvohBDd5hs)

Claude Community Ambassador at Anthropic Chile — one of 9 leaders selected globally for the Claude Impact Lab program. Builder and founder of Bendita IA (4,500+ members), Chile's most active AI community, and Zyvor, Latin America's first autonomous growth agency operated by AI agents. Architect of Agent OS — an operating system for orchestrating intelligent agents in organizations, defining how companies deploy and govern AI at scale. Focused on one thing: moving Latin America from AI consumer to AI builder.

- Event context: 🔥 AI TINKERERS - SANTIAGO / CAPÍTULO #10 / 2026 — 2026-04-16 — Santiago
- Public talk page: https://santiago.aitinkerers.org/talks/rsvp_HCvohBDd5hs

### [How to Argue With a Language Model (And Win)](https://belgium.aitinkerers.org/talks/rsvp_RV-s6edPsfU)

You've built an AI agent, it mostly works, and now you're stuck in a loop of tweaking prompts and hoping for the best. Sound familiar? In this talk we'll move past vibes-based development and into structured experimentation. We'll cover how to set up A/B tests for your agents, build and curate datasets from captured interactions or static data, and wire up evals that actually tell you whether your changes made things better or worse.

- Event context: AI Tinkerers Antwerp Meetup - April 1 (no joke) — 2026-04-01 — Belgium
- Public talk page: https://belgium.aitinkerers.org/talks/rsvp_RV-s6edPsfU

### ["Review Before Review" — the paradox of fixing code before it hits review](https://boston.aitinkerers.org/talks/rsvp_yiw0lZLHsqk)

Your AI coding tool writes hundreds of lines while you grab coffee. It just doesn't know why you banned that pattern two refactors ago, or what your team means by "clean architecture," or which module boundaries you've spent months enforcing. You find out at PR review. Sometimes you don't. ATX reads your codebase and deploys a fleet of specialized AI agents — built from your actual code, not templates — that run silently inside Claude Code and review every change against your rules in real time. One command to set up. Nothing added to your repo. Fully local. → Download ATX - https://iteration.sh/try → ATX Docs - https://iteration.sh/atx/docs/ → See it live in action - https://github.com/ric03uec/clawrium/pull/200

- Event context: AI Tinkerers Boston: March 2026 Meetup — 2026-03-30 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_yiw0lZLHsqk

### [Kill the Account](https://paris.aitinkerers.org/talks/rsvp_0EWKqOx7mrY)

AI agents are no longer limited by human-operated infrastructure. With the rise of wallet-based authentication and emerging standards like x402, agents can now directly access web search APIs without signup flows, API keys, or manual configuration. In this talk, Guillaume from Linkup will walk through how this shift fundamentally changes how we build and deploy agentic systems. He’ll present Linkup’s latest release with x402 support, showing how removing the final human bottleneck unlocks fully autonomous, composable AI workflows. The session will combine real-world demos with architectural insights, giving both technical and product perspectives on what comes next for autonomous agents on the web.

- Event context: AI Builders Meetup Paris - OpenClaw (featuring 42 AI & HEC Vibe) — 2026-03-26 — Paris
- Public talk page: https://paris.aitinkerers.org/talks/rsvp_0EWKqOx7mrY

### [How AI Agents Work, from while to wow!](https://paris.aitinkerers.org/talks/rsvp_cVHew7whhkA)

A clone of OpenCLAW to learn how it works

- Event context: AI Builders Meetup Paris - OpenClaw (featuring 42 AI & HEC Vibe) — 2026-03-26 — Paris
- Public talk page: https://paris.aitinkerers.org/talks/rsvp_cVHew7whhkA

### [AI-first GTM Playbook](https://seattle.aitinkerers.org/talks/rsvp_mSfnWcYb68k)

The singularity is now defined as human attention—distribution is more important than ever. Over the past 6 months we've been working with startups on developing unique plays for them and their business—but there is commonality in their unique patterns. So what might you experiment with? We've Open Sourced an ideation asset -- called the Tarka GTM Playbook. After spending over 100M tokens in analyzing our last 6 months of logs, we're releasing our first structured canonical playbook—for free. Browse the website as a user or agent (MCP) or browser agent (WebMCP enabled) with helpful search tools to build your unique plan and get guided instructions to build the right plays—available today.

- Event context: AI Tinkerers Seattle: GTM Engineering Kickoff — Building AI for Growth — 2026-02-26 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_mSfnWcYb68k

### [From Chatbots to Sandboxes: Why AI Needs an Execution Layer](https://singapore.aitinkerers.org/talks/rsvp_5A7Kmd5lcps)

The audience at this event is already fluent in AI hype. This talk is designed to cut through that by focusing on what actually works in practice: shipping AI systems that can reliably execute tasks in the real world. Founders, operators, and developers experimenting with agents or automations often hit the same wall. Chat-based tools are impressive, but they fall apart when tasks require memory, state, files, or time. By sharing lessons from building Mogra’s persistent sandbox, this talk gives attendees a clearer mental model for where agentic AI breaks today and how to design around those limitations.

- Event context: AI Tinkerers - The Age of AI & Infrastructure (Singapore) — 2026-02-11 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_5A7Kmd5lcps

### [Calling Tools like code programmatically](https://seattle.aitinkerers.org/talks/rsvp_Mmbrn3sy9nc)

Codecall changes how agents interact with tools by letting them write and execute code (in deno sandboxes) that orchestrates multiple tools programmatically (like an API) to do a task, rather than making individual tool calls that bloat the context and increase the token usage like in traditional agents

- Event context: AI Tinkerers Seattle: January Meetup — 2026-01-31 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_Mmbrn3sy9nc

### [Behind FreshFront’s Many AI Agents](https://toronto.aitinkerers.org/talks/rsvp_Ny1UsAT1CSM)

Guiding audience through the Commandr automation frameworks and Large Web Model FreshFront uses to automate web browsing tasks, the provisions / fallbacks we have in place to bypass captchas and securely authenticate with credentials during browser automation. I will also be demonstrating the code for FreshFront’s AI ecommerce store builder and AI workspaces.

- Event context: AI Tinkerers Toronto - December Meetup sponsored by Auth0 and TribalScale! — 2025-12-03 — Toronto
- Public talk page: https://toronto.aitinkerers.org/talks/rsvp_Ny1UsAT1CSM

### [Building a full-stack AI business co-founder](https://amsterdam.aitinkerers.org/talks/rsvp_iQHibU-m7a8)

Starnus makes the vision of a 1-person billion-dollar company achievable. Founders can rapidly build products with AI but what about running their business? Starnus simplifies marketing, sales, funding, and daily operations by integrating third-party AI agents/tools and orchestrating them within its infrastructure. Users can get complex tasks done through simple prompts. Our marketplace and pay-as-you-go model ensure simplicity, affordability, and scalability. In the talk, I want to focus on 1 or 2 of our workflows, mainly investors finder and lead generator, and how we've integrated with 3rd party AI agents/tools, and also by utilizing Gemini, we've achieved a result 10X better than any generic tool like ChatGPT.

- Event context: AI Tinkerers Amsterdam - July Edition — 2025-07-24 — Amsterdam
- Public talk page: https://amsterdam.aitinkerers.org/talks/rsvp_iQHibU-m7a8

### [Nuance](https://singapore.aitinkerers.org/talks/rsvp_WXOTat5Ick0)

Virtual Project Manager Agent that understands full context of any project and code that does your work for you

- Event context: AI Tinkerers Singapore: 6th Meetup - April 25th, 2025 — 2025-04-25 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_WXOTat5Ick0

### [Automated Recruitment Researcher - How AI Agents help with multi-factor fuzzy candidate scoring over graph data?](https://poland.aitinkerers.org/talks/rsvp_yGJaBBvmT_o)

This talk examines how AI agents can revolutionize recruitment by analyzing complex graph data and applying fuzzy logic to create more nuanced candidate scoring systems.

- Event context: AI Tinkerers Poland #3 - Meetup in Warsaw (March) — 2025-03-20 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_yGJaBBvmT_o

### [AutoTask - AI brain that knows everything you're doing on your laptop, and auto completes your routine tasks for you](https://boston.aitinkerers.org/talks/rsvp_BFg_QqfZzBw)

AutoTask analyzes the daily activity on your computer and offers to automate repetitive and boring tasks using AI agents. Are there tasks that you know you need to do but dread having to do them? Applying for jobs, shopping for groceries, paying your bills, travel arrangements, booking appointments, organizing your digital photos, etc. What if AI could take over some of those tasks, and do them on your behalf? That's where AutoTask comes in - it watches what you're doing throughout the day, and sees which tasks you're repeatedly doing, and comes up with a plan for automating those tasks. These tasks come into the AutoTask dashboard where you can review the plan before accepting the task. Once you accept the task, AutoTask gets to work - firing up a browser and executing the plan to complete the task. When it's done with the task, it reports back what it did in the dashboard.

- Event context: AI Tinkerers Boston February 2025 — 2025-02-24 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_BFg_QqfZzBw

### [alBERT](https://singapore.aitinkerers.org/talks/rsvp_01348tW6K3A)

A desktop launcher that learns from your chrome browsing activity to generate tool calls for agents to use. Essentially, alBERT is a tool that unlocks the browser for AI agents.

- Event context: AI Tinkerers Singapore: 5th Meetup - February 21st, 2025 — 2025-02-21 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_01348tW6K3A

### [Xelerit - AI Copilot for industrial robotics](https://zurich.aitinkerers.org/talks/rsvp_a35OOTyQzVs)

I will go through how our software works, which mirror the complete work of a robotics engineer, making it much faster. Our mvp has: • robot code generation (in the native robot-brand language) • copilot chat (for easy navigation of robot docs) • code translator between robot languages • I/O automatic configuration from PLC to robot. • Simulation

- Event context: AI Tinkerers Zurich - February 6 — 2025-02-06 — Zürich
- Public talk page: https://zurich.aitinkerers.org/talks/rsvp_a35OOTyQzVs

### [Self Correcting Data Extractor](https://mumbai.aitinkerers.org/talks/rsvp_Zr3pPQHK26M)

Coding agents collaboratively write and debug Python code to convert bank statements from any bank into standardized JSON. Using various PDF-to-text libraries, agents iteratively refine their code to handle errors. Successful code is stored for future processing of statements from the same bank.

- Event context: 🚀 AI Tinkerers Mumbai Chapter: Community Social Event! — 2025-01-25 — Mumbai
- Public talk page: https://mumbai.aitinkerers.org/talks/rsvp_Zr3pPQHK26M

### [Agents: Thinking Fast and Slow](https://bengaluru.aitinkerers.org/talks/rsvp_PZxSQjRYFmQ)

For AI agents to operate seamlessly across diverse applications, they must dynamically expand context to retrieve relevant information and converge on the right actions without relying on predefined workflows. This talk explores how we enable agents to adapt to new environments, effectively acting as the UI for AI to navigate and interact with software systems. We'll cover: Context Expansion: Dynamically retrieving and structuring relevant knowledge for better decision-making. Convergence on Demand: Iterative reasoning strategies that refine actions through multi-step execution. Ontology-Guided Reasoning: Leveraging structured knowledge to improve accuracy and decision reliability. Real-World Applications: How these techniques allow agents to orchestrate complex workflows, break API limitations, and drive automation across SaaS tools. By combining structured retrieval with adaptive reasoning, we move beyond static integrations—unlocking AI agents that can truly think in motion and execute with precision.

- Event context: AI Tinkerers - Bengaluru - January Meetup — 2025-01-23 — Bengaluru
- Public talk page: https://bengaluru.aitinkerers.org/talks/rsvp_PZxSQjRYFmQ

### [Vital AI Agent Ecosystem + Chat.ai](https://nyc.aitinkerers.org/talks/rsvp_EplSu1Cm9yk)

We're building an open standard for deploying A.I. Agents and for A.I. Agent collaboration. Vital AI Agent Ecosystem is an open-source reference implementation of these. We'll demo this platform for deploying an Agent and enabling different Agents to communicate and collaborate with each other and people.

- Event context: AI Tinkerers - New York City - December 2024 Meetup — 2024-12-12 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_EplSu1Cm9yk

### [Searching Millions of Tables with AI](https://nyc.aitinkerers.org/talks/rsvp_1hKK26gH1VY)

The challenges with AI data analysis mirror those of humans--the hard part is finding and prepping the data, and the analysis itself is the easy part. Current AI tech mostly relies on humans to deal with the hard part by constraining the scale to a handful of tables and requiring human curation via data warehouses, databases, and other tabular file formats. I'll demo a prototype I've built that discovers open datasets across the web & indexes them so that users (and AI agents!) can search across millions of tables/columns along dimensions like semantic similarity, joinability, data shape, data quality, etc.

- Event context: AI Tinkerers - New York City - November 2024 Meetup | Win Meta Ray-Bans — 2024-11-12 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_1hKK26gH1VY

### [GenAI Firewall - Taming GenAI in the Enterprise Environments](https://prague.aitinkerers.org/talks/rsvp_5_L52BsrcOU)

How do employees or end users interact with GenAI based chatbots? How do Murphy's laws apply? How to use AI contra AI to detect and block potential threats? We will demo our MVP/PoC of a solution dealing with the new challenges.

- Event context: AI Tinkerers Prague October Meetup - Days of AI — 2024-10-15 — Prague
- Public talk page: https://prague.aitinkerers.org/talks/rsvp_5_L52BsrcOU

### [Meal Planner](https://hamburg.aitinkerers.org/talks/rsvp_fTftzEHFUN8)

Currently trying to make a meal planner for myself by using langgraph and a combination of ai agents and chains. At the time of writing I am able to complete a short onboarding in a chat where the user is asked about preferences, dislikes and allergies etc. Those infos being extracted are then used to generate a meal plan for the next 7 days. I already have some recipes being retrieved but I want to build it further to have some sort of long term memory where you can give feedback why you don't want a recipe that was planned for you. That information should then be used to get a different recipe as well as commiting that information to long term memory. E.g. there is a recipe with avocado and you don't like avocoado - you also did not mention it before in the onboarding because you simply didn't think about it. Saying you want a different recipe because you don't like avocado.

- Event context: AI Tinkerers Hamburg - September 12 — 2024-09-12 — Hamburg
- Public talk page: https://hamburg.aitinkerers.org/talks/rsvp_fTftzEHFUN8

### [Conversational AI recruiting](https://nyc.aitinkerers.org/talks/rsvp_RniR2uzqFiY)

Demo of a project that I am building called phonescreen.ai How I’m using conversational AI to help improve and automate the candidate phone screen step in the interview process.

- Event context: AI Tinkerers August Meetup — 2024-08-28 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_RniR2uzqFiY

## Related Technologies

- [GPT-4](https://aitinkerers.org/technologies/gpt-4) ([Markdown](https://aitinkerers.org/technologies/gpt-4.md)) — 529 public demos
- [BERT](https://aitinkerers.org/technologies/bert) ([Markdown](https://aitinkerers.org/technologies/bert.md)) — 179 public demos
- [GPT-3](https://aitinkerers.org/technologies/gpt-3) ([Markdown](https://aitinkerers.org/technologies/gpt-3.md)) — 191 public demos
- [BLOOM](https://aitinkerers.org/technologies/bloom) ([Markdown](https://aitinkerers.org/technologies/bloom.md)) — 115 public demos
- [Llama-2](https://aitinkerers.org/technologies/llama-2) ([Markdown](https://aitinkerers.org/technologies/llama-2.md)) — 227 public demos
- [LLM](https://aitinkerers.org/technologies/llm) ([Markdown](https://aitinkerers.org/technologies/llm.md)) — 123 public demos
- [PaLM 2](https://aitinkerers.org/technologies/palm-2) ([Markdown](https://aitinkerers.org/technologies/palm-2.md)) — 116 public demos
- [RoBERTa](https://aitinkerers.org/technologies/roberta) ([Markdown](https://aitinkerers.org/technologies/roberta.md)) — 118 public demos
- [AI](https://aitinkerers.org/technologies/ai) ([Markdown](https://aitinkerers.org/technologies/ai.md)) — 55 public demos
- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 215 public demos
- [APIs](https://aitinkerers.org/technologies/apis) ([Markdown](https://aitinkerers.org/technologies/apis.md)) — 19 public demos
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 74 public demos
- [GraphRAG](https://aitinkerers.org/technologies/graphrag) ([Markdown](https://aitinkerers.org/technologies/graphrag.md)) — 13 public demos
- [Keras](https://aitinkerers.org/technologies/keras) ([Markdown](https://aitinkerers.org/technologies/keras.md)) — 74 public demos
- [MCP](https://aitinkerers.org/technologies/mcp) ([Markdown](https://aitinkerers.org/technologies/mcp.md)) — 129 public demos
- [ONNX](https://aitinkerers.org/technologies/onnx) ([Markdown](https://aitinkerers.org/technologies/onnx.md)) — 83 public demos
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [PyTorch](https://aitinkerers.org/technologies/pytorch) ([Markdown](https://aitinkerers.org/technologies/pytorch.md)) — 273 public demos

## More Results

- Next: https://aitinkerers.org/technologies/ai-agents.md?page=2