# Pydantic Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/pydantic
> Markdown URL: https://aitinkerers.org/technologies/pydantic.md
> Technology record last updated: 2026-09-18T15:13:51Z
> Generated: 2026-09-23T02:33:19Z

Pydantic is Python's most-used data validation library: it enforces data schemas using standard type hints and boasts a Rust-core for exceptional speed.

Pydantic is the premier data validation and parsing library for Python. It mandates data structure using pure, canonical Python type annotations, drastically reducing boilerplate code. With over 360M monthly downloads, Pydantic is battle-tested: all FAANG companies and major frameworks (FastAPI, SQLModel, LangChain) rely on it for robust data handling. Its core validation logic is written in Rust, ensuring high performance. Pydantic models also generate JSON Schema, facilitating seamless integration and documentation for API development.

- Official technology site: https://pydantic.dev
- Public AI Tinkerers demos and talks: 30
- Result page: 1 of 2

## Recent Public Talks and Demos

### [Dev tools are all you need](https://seattle.aitinkerers.org/talks/rsvp_FBwC1XbQpYE)

Maibook is a desktop app - a network of you and AI agents personalized based on your file system activity. Demo will show: - Maibook briefly - Customized dev tools built for the project - strings_web - centrally manage all strings AI agents see - system and user prompts, tool schemas, tool docstrings, tool field descriptions and constraints - edit in tool, updates source files - dev_web - see sessions and debug what works, what doesn't - inspired by cc viewer - in the works - ccdbug - Others - threadviewer, benchmarking models, etc.

- Event context: AI Tinkerers Seattle: May Demo Night (Ft. Oracle & NVIDIA) — 2026-05-22 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_FBwC1XbQpYE

### [Patent Mining for Engineers: Building an Agentic RAG System for Inventive Problem Solving using TRIZ & AI](https://poland.aitinkerers.org/talks/rsvp_oYsjqZvaY7E)

A pipeline that parses patent PDFs, extracts Technical Contradictions, classifies solutions into TRIZ Inventive Principles, and indexes everything into a vector database. This collection is then feeding an AI Agent that helps engineers solve real inventive problems. The demo starts with a raw patent PDF, submits it live to the processing endpoint, and walks through what gets extracted and indexed. Then, given a real mechanical engineering problem, the agent frames it as a TRIZ contradiction, retrieves relevant patents, and proposes concrete solution ideas, powered by domain knowledge, not just plain LLM generation.

- Event context: AI Tinkerers Poland #3 - Meetup in Wrocław — 2026-05-06 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_oYsjqZvaY7E

### [predict-rlm: Harness-less Recursive Language Models Built on DSPy](https://montreal.aitinkerers.org/talks/rsvp_xSr_Hb9LQQw)

predict-rlm is a callable RLM (Recursive Language Model) runtime built on top of DSPy. You define your inputs, outputs, and tools — the model handles its own control flow via a sandboxed REPL. I'll demo how it processes documents (PDFs, images, invoices) end-to-end, showing how a single RLM call can recursively decompose complex tasks into sub-LM calls — with fully interpretable trajectories and structured outputs — all without context rot.

- Event context: AI Tinkerers Montreal - April Demo Night — 2026-04-22 — Montreal
- Public talk page: https://montreal.aitinkerers.org/talks/rsvp_xSr_Hb9LQQw

### [From Chat to CASE: Building an AI Agent That Scouts and Fills Real Complaint Forms](https://singapore.aitinkerers.org/talks/rsvp_y-Gl78fWqSw)

I built Onestop, a chat-first complaint agent that turns a user’s issue into an execution workflow: it extracts the complaint from natural language, plans the escalation path, drafts the complaint email, and for verified portals like CASE Singapore, scouts the live web form and attempts a browser fill through a Form Genome pipeline. The demo is intentionally technical and code-first: a FastAPI orchestrator, structured complaint state, Playwright-based browser execution, and truthful fallbacks when a route is not production-safe.

- Event context: AI Tinkerers Singapore: The Agentic Future & Dev/Eng Workflows — 2026-04-21 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_y-Gl78fWqSw

### [A language of logic](https://cologne.aitinkerers.org/talks/rsvp_E9bB1JtSMsU)

We're integrating LLMs with mathematical proof solvers/constrain optimizers to allow our LLM driven system to carry out specific logical tasks and calculations while maintaining explainability for the user.

- Event context: AI Tinkerers Cologne 4: Live Technical Demos — 2026-04-16 — Cologne
- Public talk page: https://cologne.aitinkerers.org/talks/rsvp_E9bB1JtSMsU

### [Wisdom Layer: Agent Cognitive Architecture](https://manchester-nh.aitinkerers.org/talks/rsvp_hAZ9fGsxhsI)

The Wisdom Layer is a Python SDK that wraps any LLM with persistent memory that matures, autonomous reflection cycles, self-authored behavioral rules with a lifecycle, and an internal critic that evaluates output against active rules before it ships. The demo at AI Tinkerers ran on Claude Haiku — Anthropic's cheapest model — and showed how a commodity model becomes substantially more reliable when scaffolded with memory, reflection, and enforcement, rather than fine-tuned. Early benchmark showed fabrication rate dropping from 22% to 2% on the same model (n=45 single corpus, broader eval in progress). The same architecture also powers loom-code (AI-assisted coding across 20+ repos) and a computational pharmacogenomics research platform in active collaboration discussions with academic cancer centers.

- Event context: AI Tinkerers Manchester (Bedford), NH - April 2026 Meetup — 2026-04-15 — Manchester NH
- Public talk page: https://manchester-nh.aitinkerers.org/talks/rsvp_hAZ9fGsxhsI

### [Turning Claude Code into a Self-Improving Agentic Development Platform](https://orange-county.aitinkerers.org/talks/rsvp_iO2rjPLhMOE)

Every AI coding session is a chain-of-thought reasoning trace — step-by-step problem solving with verifiable outcomes. BashGym captures these traces and uses them to train a reasoning language model with the same techniques behind frontier RLMs: GRPO for reinforcement learning, RLVR for verifiable reward signals from test results, and distillation to transfer reasoning from a large teacher into a small local model. The result is a personal RLM trained on how you actually think through code — your conventions, your repos, your patterns.

- Event context: AI Tinkerers OC - March 11, 2026 Meetup — 2026-03-12 — Orange County
- Public talk page: https://orange-county.aitinkerers.org/talks/rsvp_iO2rjPLhMOE

### [Building for the Future Internet: Agent Services](https://upstate-ny.aitinkerers.org/talks/rsvp_GetoIsT3wio)

meetSpace - exploring an agent-first service for creating and finding local IRL events.

- Event context: AI Tinkerers Upstate NY: March Meetup — 2026-03-10 — Upstate NY
- Public talk page: https://upstate-ny.aitinkerers.org/talks/rsvp_GetoIsT3wio

### [The Semantic Kill Chain: Bypassing Deterministic Memory with Malicious Intents](https://toronto.aitinkerers.org/talks/rsvp_8mrzc9297Xw)

A live demonstration of "Semantic Laundering" in Agentic Memory. I will show how LLM extraction pipelines can be coerced into acting as proxies for malicious commands. By wrapping destructive instructions in the semantic texture of valid ontology entities, I will demonstrate how standard structural validation fails. The system validates syntax but remains blind to semantics, committing the payload to the Knowledge Graph as an actionable tool.

- Event context: AI Tinkerers Toronto - February 2026 @ Cohere! — 2026-02-26 — Toronto
- Public talk page: https://toronto.aitinkerers.org/talks/rsvp_8mrzc9297Xw

### [Building a Production-Ready ESG KPI Extraction Pipeline with Gemini API](https://vienna.aitinkerers.org/talks/rsvp_LPKQctsj3Mw)

A practical deep-dive into building an AI-powered system that extracts 170+ structured KPIs from ESG/financial PDF documents. I'll walk through the real engineering decisions behind our two-stage extraction pipeline: how we use Gemini's Files API with explicit caching to control costs, parallel structured outputs processing to speed up extraction, LLM-based conflict resolution for multi-document scenarios and how we evaluate the pipeline. Expect code snippets, architecture diagrams, and honest lessons learned from development.

- Event context: AI Tinkerers Vienna: 2026 — 2026-02-19 — Vienna
- Public talk page: https://vienna.aitinkerers.org/talks/rsvp_LPKQctsj3Mw

### [Structured charting with LLMs](https://dublin.aitinkerers.org/talks/rsvp_khbV0Xbmnds)

LLMs are great for producing charts however asking for the same chart style 100 times in a row, results are guaranteed to be inconsistent. This is where Vizdantic appears a simple lightweight yet extremely useful charting import which allows for consistent branding and smart chart type routing

- Event context: AI Tinkerers Dublin Meetup — Baseline, January 26, 2026 — 2026-01-26 — Dublin
- Public talk page: https://dublin.aitinkerers.org/talks/rsvp_khbV0Xbmnds

### [I'm in Danger - Ralph Wiggum in Claude Code](https://dublin.aitinkerers.org/talks/rsvp_mWOPI8K1sho)

I have recently been working on the idea of using an orchestrator agent to manage Claude Code in order to decrease the amount of time waiting for the coding agent to go through the building cycle. Also looking at optimisations such as using TOON (Token-Oriented Object Notation) file types to also optimise token usage.

- Event context: AI Tinkerers Dublin Meetup — Baseline, January 26, 2026 — 2026-01-26 — Dublin
- Public talk page: https://dublin.aitinkerers.org/talks/rsvp_mWOPI8K1sho

### [ThailandDev: AI Powered Job Board Pipeline](https://chicago.aitinkerers.org/talks/rsvp_BMSed1pAHJ0)

ThailandDev is a job board for tech roles in Thailand, powered by an AI pipeline that requires zero manual scraping rules. I'll walk through how we use Gemini 2.5 Flash to analyze any career page structure, extract job URLs, and pull structured data including skills, seniority level, remote policy, and language requirements. The talk covers the full stack: Crawl4AI for browser automation, Pydantic for schema validation, Supabase for storage, GCP Cloud Run for scheduling, and Next.js for the frontend. Adding a new company to scrape? Just drop in a URL.

- Event context: AI Tinkerers Chicago January Meetup ft. PostHog — 2026-01-20 — Chicago
- Public talk page: https://chicago.aitinkerers.org/talks/rsvp_BMSed1pAHJ0

### [MonoMorph: An Agentic tool for refactoring monoliths into microservices](https://montreal.aitinkerers.org/talks/rsvp_l2DEga8TIMc)

I'll be doing a live demo of MonoMorph, an agentic system built on LangGraph that automatically refactors monoliths into microservices. The talk will focus on the agent workflows we designed and the key lessons we learned during its development. For example, to guarantee the correctness of the refactored code, we needed to heavily ground our code-generation agent with structured outputs and template driven prompts to ensure it fits the target design. However, our bug-fixing agent required a much more flexible, multi-turn method to autonomously debug unexpected compilation errors in a Docker-based loop. During the presentation, we’ll look into the design of the workflows for each of the agents, examples of the templates we used and the implementations of the tools that are critical to the success of the agents. We’ll run a demo of the application which we will keep checking on as we discuss the rest.

- Event context: AI Tinkerers Montreal: Demo Night — November 20, 2025 — 2025-11-20 — Montreal
- Public talk page: https://montreal.aitinkerers.org/talks/rsvp_l2DEga8TIMc

### [Automating error analysis - what works and what doesn't](https://liverpool.aitinkerers.org/talks/rsvp_9LjshIDL1Do)

Error analysis is known to be the highest ROI step in building reliable agents that work. However, as agent & task complexity grow, manual approaches to it become prohibitive, and naive ways of automating it remain unsatisfying. How *should* developers scale up this key step in agent evaluation? Here, I'll present our recent findings on what works - and what doesn't - when automating error analysis. TL;DR: we can do a lot better than stuffing everything into an LLM, with a bit of intentional context engineering and task decomposition.

- Event context: AI Tinkerers Liverpool HITL Meetup at FACT Lab — 2025-10-28 — Liverpool
- Public talk page: https://liverpool.aitinkerers.org/talks/rsvp_9LjshIDL1Do

### [Building a Custom Machine Learning Solution for Sensitive Information Detection](https://montreal.aitinkerers.org/talks/rsvp_ySKYULBmt78)

Please watch the 5 minutes demo

- Event context: AI Tinkerers Montreal – Spooky October 2025 Meetup — 2025-10-21 — Montreal
- Public talk page: https://montreal.aitinkerers.org/talks/rsvp_ySKYULBmt78

### [Test2Synth: controlling hardware music geart from text](https://tokyo.aitinkerers.org/talks/rsvp_gicqjGFQ-wc)

A dirty attempt at interfacing LLM and test-based interface with music hardware. Hardware synthesizer have seen a resurgence. They can be fairly complex to build. We will show how to build a simple LLM-based system to program synthesizer from text input. By interfacing LLM to hardware synth through MIDI, the underlying principle can in theory control any MIDI-enabled system. Depending on the A/V capabilities at the demo, I will have either a video to present, or demonstrate live with actual hardware how to make sound from the system.

- Event context: AI Tinkerers Tokyo Kickoff – October 10, 2025 — 2025-10-10 — Tokyo
- Public talk page: https://tokyo.aitinkerers.org/talks/rsvp_gicqjGFQ-wc

### [Shipping AI for Good with LLM-as-Judge Evals](https://nyc.aitinkerers.org/talks/rsvp_Iv0GfB9IJjk)

Immigrants facing detainment need urgent, clear answers but face dense legal docs. This critical gap inspired me to build a tool to provide immediate support in a high-stakes situation. I built a bilingual (EN/ES) RAG chatbot that provides instant, conversational answers to immigration questions, drawing from trusted non-profit legal FAQs.

- Event context: NYC October Demo Day — 2025-10-02 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_Iv0GfB9IJjk

### [Agent Builds Production AI Workflow in 4 Minutes](https://paris.aitinkerers.org/talks/rsvp__BV1raa1xGM)

I'll demonstrate an agent transforming natural language requirements into a production-ready AI workflow using our open-source declarative language for AI automation. We could call this Vibe-Coding a workflow or Vibe-Flowing. **The Live Demo** - Start with a business requirement such as expense report processing: *"Extract employee info and line items, process them one by one, validate against company rules, flag anomalies and write rejection email"* - Watch the agent design a multi-step pipeline with specialized pipes for each transformation - See it validate the workflow statically, then with a dry run on mock data, then with synthetic test cases - Deploy the final pipeline that handles document variations while guaranteeing structured outputs **What I'll Show Under the Hood** - The declarative TOML syntax that captures business intent, not API calls - How each pipe declares its conceptual contract: `PurchaseDocument → Buyer` instead of `string → JSON` - The agent's reasoning as it decomposes the mega-task into specialized steps - How it selects different models for different pipes: OCR, Vision, LLM, Image generation - The testing loop where the agent validates outputs and adjusts prompts **The Technical Architecture** - **Declarative abstraction layer**: Define what transformations you want, not how to implement them - **Deterministic structure with adaptive execution**: Generate validated Pydantic BaseModels at every run, while LLMs handle content variation - **Compositional design**: Pipes flow sequentially, batch in parallel, branch conditionally - **Agent-native format**: workflow definition that reads like documentation, perfect for LLM generation and refinement

- Event context: Epic Demo night 0.2.1 - Sep 18th, 2025 — 2025-09-18 — Paris
- Public talk page: https://paris.aitinkerers.org/talks/rsvp__BV1raa1xGM

### [Claudio AI Musician](https://milan.aitinkerers.org/talks/rsvp_lxUiuQMNPYs)

A short demonstration of AI Agents for music production, the agents collaborate with the user to control a FM Synthesiser and jam together to create a dataset that will be used to train a more personalised AI tailored to the users musical tastes and needs.

- Event context: AI Tinkerers Milan - June 10, 2025 - Community Demos & Networking — 2025-06-10 — Milan
- Public talk page: https://milan.aitinkerers.org/talks/rsvp_lxUiuQMNPYs

### [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

### [School Quiz Grader with DocRouter.AI](https://boston.aitinkerers.org/talks/rsvp_9vvWQ2XeaSc)

DocRouter.AI can serve as the AI tech stack backbone for companies in HealthTech, Accounting, Legal, Insurance verticals dealing with unstructured documents. You customize your application as needed for your use case - while relying on DocRouter.AI for full power of LLM prompt engineering and human-in-the-loop as back end. In this demo, we have an EdTech application: An AI evaluation tool for 5th grade quizzes, with open form items (no multiple choice!), with the teacher-in-the-loop.

- Event context: AI Tinkerers Boston May Meetup — 2025-05-20 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_9vvWQ2XeaSc

### [Large Scale Narrative Analysis via Structured Generation](https://milan.aitinkerers.org/talks/rsvp_Ed6-N6ff6ZQ)

We present a large scale pipeline for creating Entity Interaction Networks from English novels and show how this meso-structure reveals a natural taxonomy of novels across time and genres.

- Event context: AI Tinkerers Milan - May 8, 2025 — 2025-05-08 — Milan
- Public talk page: https://milan.aitinkerers.org/talks/rsvp_Ed6-N6ff6ZQ

### [Tactician: Detailed, specific workplace advice for confusing or political situations](https://seattle.aitinkerers.org/talks/rsvp_gumR4iFgXLM)

I'm baking everything I learned through my PhD in Organizational Behavior and my years as a professor in the field, combined with my lifelong interest in organizational effectiveness, into Tactician. Tactician isn't a single advice-bot. It's a system that combines distinctly different perspectives, careful questioning and hypothesis-testing, and freeform chat to get you evidence-based, actionable advice. In the live demo I'll walk through a real workpalce scenario (people come to me for advice with these all the time). We'll chat with the advisor personas first. Then we'll give some further tailored details, based on an intake survey that's dynamically generated based on the chat. We'll see the combination of these piped back to the advisors, who will suggest little hypothesis tests in the real world so you can figure out what is actually going on. We'll simulate this in the demo. Once we're sure of the facts, the advisors present courses of action that seem likely to lead to a integrated good outcome: something that will be beneficial both for you and the organization. It's a system based on careful data selection, an agentic workflow (BAML, pydantic), RAG from a vector db with only the choicest organizational texts in it, and (eventually) fully local and private models to offer a complete tailored advice system. Starting by hooking it up to Claude 3.7 sonnet with some prompt engineering for multiple advice personas. Use-case spiel: Maybe you interacted with a boss or client recently who said something confusing but important. Maybe you're in a situation where people expect you to have a lot of people skills and you feel overwhelmed. Use Tactician to figure out a) what's going on, b) what your options are to get to a good outcome.

- Event context: AI Tinkerers Seattle - April Meetup — 2025-04-25 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_gumR4iFgXLM

## Related Technologies

- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 664 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [FastAPI](https://aitinkerers.org/technologies/fastapi) ([Markdown](https://aitinkerers.org/technologies/fastapi.md)) — 181 public demos
- [LangChain](https://aitinkerers.org/technologies/langchain) ([Markdown](https://aitinkerers.org/technologies/langchain.md)) — 446 public demos
- [Next](https://aitinkerers.org/technologies/next) ([Markdown](https://aitinkerers.org/technologies/next.md)) — 186 public demos
- [JSON Schema](https://aitinkerers.org/technologies/json-schema) ([Markdown](https://aitinkerers.org/technologies/json-schema.md)) — 9 public demos
- [RAG](https://aitinkerers.org/technologies/rag) ([Markdown](https://aitinkerers.org/technologies/rag.md)) — 147 public demos
- [Anthropic API](https://aitinkerers.org/technologies/anthropic-api) ([Markdown](https://aitinkerers.org/technologies/anthropic-api.md)) — 66 public demos
- [Claude](https://aitinkerers.org/technologies/claude) ([Markdown](https://aitinkerers.org/technologies/claude.md)) — 174 public demos
- [claude-4](https://aitinkerers.org/technologies/claude-4) ([Markdown](https://aitinkerers.org/technologies/claude-4.md)) — 8 public demos
- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 216 public demos
- [DSPY](https://aitinkerers.org/technologies/dspy) ([Markdown](https://aitinkerers.org/technologies/dspy.md)) — 12 public demos
- [Gemini](https://aitinkerers.org/technologies/gemini) ([Markdown](https://aitinkerers.org/technologies/gemini.md)) — 188 public demos
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 74 public demos
- [GPT-4](https://aitinkerers.org/technologies/gpt-4) ([Markdown](https://aitinkerers.org/technologies/gpt-4.md)) — 529 public demos
- [LiteLLM](https://aitinkerers.org/technologies/litellm) ([Markdown](https://aitinkerers.org/technologies/litellm.md)) — 21 public demos
- [LLMs](https://aitinkerers.org/technologies/llms) ([Markdown](https://aitinkerers.org/technologies/llms.md)) — 83 public demos
- [OpenAI](https://aitinkerers.org/technologies/openai) ([Markdown](https://aitinkerers.org/technologies/openai.md)) — 112 public demos

## More Results

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