# Langfuse Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/langfuse
> Markdown URL: https://aitinkerers.org/technologies/langfuse.md
> Technology record last updated: 2026-09-18T15:13:48Z
> Generated: 2026-09-22T08:36:39Z

Langfuse is the open-source LLM engineering platform: gain full observability, manage prompt versions, and run production-grade evaluations.

Langfuse delivers the essential LLM engineering stack. It's an open-source platform for full-lifecycle management of your AI applications (agents, chains, etc.). Use its comprehensive tracing (OpenTelemetry-based) to debug complex, non-deterministic interactions and track exact cost/latency metrics. The system provides robust prompt management (versioning, A/B testing) and flexible evaluation tools to measure output quality and monitor production health. Integrations are native: connect with Langchain, OpenAI, and LlamaIndex via Python/JS SDKs for immediate control and clarity over your LLM deployment.

- Official technology site: https://langfuse.com/
- Public AI Tinkerers demos and talks: 13
- Result page: 1 of 1

## Recent Public Talks and Demos

### [ibaAgent: Agentic time-series analysis.](https://nurnberg.aitinkerers.org/talks/rsvp_DcXjk1ibxVU)

I developed an agentic system using LangGraph to analyze high-resolution time-series data via natural language. Since high-resolution data is too token-intensive for LLMs, the agent operates "blind" by dynamically writing and executing Python analysis scripts to process data locally. This is supported by a specialized memory architecture that allows the agent to iteratively build an internal map of undocumented, messy data stores, and a rigorous Langfuse experiment pipeline used to tune hyperparameters and prompts for predictable reasoning and precise tool use. Currently transitioning from proof-of-concept to product at iba AG, my demo will showcase LangGraph traces of the agent’s execution and its self-correction within the Python environment. I will also highlight our Langfuse evaluation workflow and how the memory layer enables the agent to navigate complex data structures over time. I’m looking to exchange ideas with fellow tinkerers specifically on memory persistence, planning strategies, and robust evaluation for agentic workflows.

- Event context: AI Tinkerers Nürnberg: April Meetup — 2026-04-22 — Nürnberg
- Public talk page: https://nurnberg.aitinkerers.org/talks/rsvp_DcXjk1ibxVU

### [Practical LLM Integration: Lessons from Building a SEC Filing Analysis Project](https://houston.aitinkerers.org/talks/rsvp_Vt9RZstb_VY)

A live coding demo walking through an LLM-powered document analysis system. I'll run the pipeline live and show the code that makes it work. Using regex and rule-based filters to reduce LLM API costs by 80%+ Extracting structured data from unstructured documents with Claude Implementing semantic caching with Qdrant to avoid redundant API calls Tracking LLM costs and latency with Langfuse

- Event context: AI Tinkerers Houston: Working Code, Live Demos, and Architectures for 2026 — 2026-01-13 — Houston
- Public talk page: https://houston.aitinkerers.org/talks/rsvp_Vt9RZstb_VY

### [Running a local LLM with guardrails](https://hong-kong.aitinkerers.org/talks/rsvp_C87wbAjEP3o)

started building a local Librechat with SSO, auditing capabilities, and guardrails.

- Event context: AI Tinkerers Hong Kong Meetup - November 27th — 2025-11-27 — Hong Kong
- Public talk page: https://hong-kong.aitinkerers.org/talks/rsvp_C87wbAjEP3o

### [Agentic AI Evaluation](https://singapore.aitinkerers.org/talks/rsvp_WUfjRFh_OmY)

Agentic AI Evaluation Agentic AI refers to systems that can plan, decide, and act autonomously across multiple steps. Evaluating such systems is harder than testing traditional AI because their behavior changes with context. Standard benchmarks often fail to capture real-world complexity and tool use. New methods combine task success rates, reasoning quality, and adaptability measures. Human-centric factors like safety, transparency, and ethics are equally important. Industry tools now provide multi-dimensional evaluation dashboards for agents. Governance and monitoring are critical for safe deployment. This talk will present key evaluation dimensions and emerging best practices. Attendees will learn how to balance performance metrics with trust and accountability. The goal is to make agentic AI both effective and responsible in real-world use.

- Event context: AIT Singapore: scalable agentic workflows - 12th August 2025 — 2025-08-12 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_WUfjRFh_OmY

### [HackerNews.coffee - Transparent AI personalisation using fast models and lean architecture](https://london.aitinkerers.org/talks/rsvp_CYF13tmV5i4)

We built a lean AI recommendation system with transparent user preference profiling for Hacker News. The key AI engineering challenge: real-time personalisation bootstrapped from minimal user data - people don't want to wait for content, but you still need quality recommendations and it all has to be transparent (and editable) for the user. This talk covers our technical approach, tradeoffs, and extending this to full content ranking.

- Event context: AI Tinkerers London Meetup - July 16th, 2025 — 2025-07-16 — London
- Public talk page: https://london.aitinkerers.org/talks/rsvp_CYF13tmV5i4

### [CTF-Agent: how to hack your site with reasoning models](https://amsterdam.aitinkerers.org/talks/rsvp_BI82WZQHbPs)

An AI agent that completes CTF (capture-the-flag) challenges using reasoning models, an agent loop and only a single tool: HTTP requests. It successfully does SQL injection and several cookie spoofing attacks of increasing complexity.

- Event context: AI Tinkerers x AdvancedJS Amsterdam — 2025-06-25 — Amsterdam
- Public talk page: https://amsterdam.aitinkerers.org/talks/rsvp_BI82WZQHbPs

### [CoffeeInABit](https://sf.aitinkerers.org/talks/rsvp__7UB3UDwzbM)

Management of LinkedIn Campaigns, which runs locally on your laptop using MCP

- Event context: AI Tinkerers - Advanced MCP Showcase — 2025-06-19 — San Francisco
- Public talk page: https://sf.aitinkerers.org/talks/rsvp__7UB3UDwzbM

### [State of browser use Agents](https://poland.aitinkerers.org/talks/rsvp_RyGQ4-bJAA4)

I've spent $ 5000 on LLM tokens for browser-use, the agent for using web browsers. Some tasks included signing up to 20k different newsletter forms (for an aggregated AI newsletter project) or gathering data (kind of like a general purpose, universal scraper). I'll show how the agent interacts with websites it hasn't seen before. I would also love to present how the agent prompt is structured by peeking inside Langfuse, and discuss prompting patterns with the community.

- Event context: AI Tinkerers Poland #3 - Meetup in Warsaw (March) — 2025-03-20 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_RyGQ4-bJAA4

### [How to direct twitch streams with a multi-agent cinema crew](https://montreal.aitinkerers.org/talks/rsvp_8m3Y7xD9Pwc)

I wanted to build a twitch streamer in a 3d environment that I could control with AI. This led me to develop a multi agent system in langgraph consisting of a director, writer, cinematographer, animation tech and audio engineer. All working together to build 1-2 minute screenplays that can be chained.

- Event context: AI Tinkerers Montreal - March 2025 Meetup at Shopify — 2025-03-12 — Montreal
- Public talk page: https://montreal.aitinkerers.org/talks/rsvp_8m3Y7xD9Pwc

### [Never Fill Forms again with "Browser Agents"](https://atlanta.aitinkerers.org/talks/rsvp_PXh82rhmgBw)

I’ll demonstrate a Chrome extension that uses an LLM driven ‘browser agent’ to navigate to a website, locate a job application form, and automatically fill out the fields with high accuracy. During the live demo, I’ll type a command like ‘Navigate to Macy’s careers page and apply for a software engineering job,’ and the browser agent will programmatically follow links, parse the relevant form, and fill out the details in real time. I’ll walk through the core technical components, including some of the code and prompts behind the automated navigation and form-filling logic, and show how the AI integrates with the extension.

- Event context: AI Tinkerers Atlanta February Meetup — 2025-02-27 — Atlanta
- Public talk page: https://atlanta.aitinkerers.org/talks/rsvp_PXh82rhmgBw

### [OpenTelemetry based instrumentation of the Vercel AI SDK (open source)](https://berlin.aitinkerers.org/talks/rsvp_TOqT2xRNW8I)

Lars (Vercel) added OTel based instrumentation to the Vercel AI SDK, we (Langfuse) built an OSS collector for it to be able to monitor complex frontend applications that use the AI SDK.

- Event context: AI Tinkerers Berlin - August 22 — 2024-08-22 — Berlin
- Public talk page: https://berlin.aitinkerers.org/talks/rsvp_TOqT2xRNW8I

### [Keeping an "AI" on LLMs with Langfuse](https://london.aitinkerers.org/talks/rsvp_Nc7KY-wgPWg)

As we've been building with LLMs at nPlan, we've often struggled to have good insight into what's happening when we deploy LLMs in the wild. Our client data is quite sensitive so I set out to explore solutions we could self-host and Langfuse was chosen as our LLM observability tool! It gives us fantastic visibility into all our user's questions, what inputs go into LLMs, what data is fetched, and much more. In this talk, I would chat through my experience setting up Langfuse and the benefits we've gotten from LLM observability already.

- Event context: AI Tinkerers London - July Meetup — 2024-07-09 — London
- Public talk page: https://london.aitinkerers.org/talks/rsvp_Nc7KY-wgPWg

### [Q&A Chatbot (Langfuse Docs)](https://berlin.aitinkerers.org/talks/rsvp_ziwzF7Fao8I)

Q&A chatbot that I've built for the Langfuse docs using pgvector, edge functions, GH actions embedding pipeline for mdx documentation, Openai streaming, the Vercel AI SDK, shadcn frontend components, and Langfuse (open-source analytics to understand what does not work and how to improve it).

- Event context: AI Tinkerers Berlin - November 24 — 2023-11-24 — Berlin
- Public talk page: https://berlin.aitinkerers.org/talks/rsvp_ziwzF7Fao8I

## Related Technologies

- [Vercel AI SDK](https://aitinkerers.org/technologies/vercel-ai-sdk) ([Markdown](https://aitinkerers.org/technologies/vercel-ai-sdk.md)) — 32 public demos
- [Gemini](https://aitinkerers.org/technologies/gemini) ([Markdown](https://aitinkerers.org/technologies/gemini.md)) — 188 public demos
- [LangChain](https://aitinkerers.org/technologies/langchain) ([Markdown](https://aitinkerers.org/technologies/langchain.md)) — 445 public demos
- [LangGraph](https://aitinkerers.org/technologies/langgraph) ([Markdown](https://aitinkerers.org/technologies/langgraph.md)) — 67 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [Vercel](https://aitinkerers.org/technologies/vercel) ([Markdown](https://aitinkerers.org/technologies/vercel.md)) — 48 public demos
- [Agent](https://aitinkerers.org/technologies/agent) ([Markdown](https://aitinkerers.org/technologies/agent.md)) — 12 public demos
- [Anthropic](https://aitinkerers.org/technologies/anthropic) ([Markdown](https://aitinkerers.org/technologies/anthropic.md)) — 36 public demos
- [Benchmark dataset](https://aitinkerers.org/technologies/benchmark-dataset) ([Markdown](https://aitinkerers.org/technologies/benchmark-dataset.md)) — 1 public demo
- [Browser](https://aitinkerers.org/technologies/browser) ([Markdown](https://aitinkerers.org/technologies/browser.md)) — 5 public demos
- [Claude](https://aitinkerers.org/technologies/claude) ([Markdown](https://aitinkerers.org/technologies/claude.md)) — 174 public demos
- [Claude Opus](https://aitinkerers.org/technologies/claude-opus) ([Markdown](https://aitinkerers.org/technologies/claude-opus.md)) — 18 public demos
- [Docker](https://aitinkerers.org/technologies/docker) ([Markdown](https://aitinkerers.org/technologies/docker.md)) — 147 public demos
- [Evals](https://aitinkerers.org/technologies/evals) ([Markdown](https://aitinkerers.org/technologies/evals.md)) — 4 public demos
- [GitHub Actions](https://aitinkerers.org/technologies/github-actions) ([Markdown](https://aitinkerers.org/technologies/github-actions.md)) — 20 public demos
- [Google Cloud Platform](https://aitinkerers.org/technologies/google-cloud-platform) ([Markdown](https://aitinkerers.org/technologies/google-cloud-platform.md)) — 25 public demos
- [GSAP](https://aitinkerers.org/technologies/gsap) ([Markdown](https://aitinkerers.org/technologies/gsap.md)) — 4 public demos