# bash Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/bash
> Markdown URL: https://aitinkerers.org/technologies/bash.md
> Technology record last updated: 2026-02-25T06:02:08Z
> Generated: 2026-08-24T00:14:35Z

Bash (Bourne-Again SHell): The ubiquitous command language interpreter and scripting engine for GNU/Linux and macOS systems.

Bash is the GNU Project's powerful shell, serving as the default command interpreter for most Linux distributions and a widely used option on macOS. Developed by Brian Fox in 1989, its name is an acronym for 'Bourne-Again SHell,' signifying its compatibility with the original Bourne Shell (sh) while incorporating advanced features from shells like ksh and csh. The technology provides a robust environment for interactive command execution and non-interactive shell scripting, supporting critical functions like command-line editing, job control, and complex programming constructs (loops, conditionals, functions) for system automation.

- Official technology site: https://www.gnu.org/software/bash/
- Public AI Tinkerers demos and talks: 12
- Result page: 1 of 1

## Recent Public Talks and Demos

### [What Is Your Coding Agent Actually Searching?](https://columbus.aitinkerers.org/talks/rsvp_t4IMPq7lJnc)

I built SearchBench, a harness for running controlled evaluation rounds over coding-agent search behavior. A SearchBench round takes real bug-localization tasks, gives an agent access to a repository, and checks whether it found the files that were actually changed in the human fix. The harness compares an incumbent search strategy against one or more challengers, then writes a static evidence bundle with exact-hit, hop-distance, token-usage, failure, and report artifacts. For the demo, I’ll run a small live ablation round over three cases. I’ll reuse cached Bash/native-search results as the incumbent, then run a few IC challenger policies in parallel. The question is: when IC succeeds or fails, was the important factor anchor quality or graph lookahead? I’ll show the actual workflow: preflight cost prediction, run planning, parallel execution, actual spend, cost-prediction error, generated bundle artifacts, and a short report explaining what changed. One thing I want to show explicitly is how much information the harness can derive from a small, bounded run: exact hits, hop distance, token usage, failure modes, and the next optimization target.

- Event context: AI Tinkerers - Columbus June Meetup — 2026-06-01 — Columbus
- Public talk page: https://columbus.aitinkerers.org/talks/rsvp_t4IMPq7lJnc

### [Scaling RAG: Hybrid Search and Hierarchical Chunking for 780k Pages](https://poland.aitinkerers.org/talks/rsvp_BCaEvuBCHLM)

I built a custom desktop-server search engine designed to help me instantly find and manage documents within my 40GB PDF library. Technical Overview: - The Interface: A Windows application where I can search and browse through the results easily. - The Search Brain: A backend powered by FastAPI that uses "hybrid search" - Data Processing: Python and Bash scripts that handle the heavy lifting, such as pulling Markdown and generating page thumbnails from every file. - Annotation AI: vLLM based LLM server that extract metadata. - The Future: I am currently adding a RAG (Retrieval-Augmented Generation) feature so I can ask the AI questions directly about the content of my documents.

- Event context: AI Tinkerers Poland - Meetup in Gdańsk #1 — 2026-04-23 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_BCaEvuBCHLM

### [Words to World, AI learning strategies for building world models.](https://san-diego.aitinkerers.org/talks/rsvp_mnJaK0Bd4l8)

Can a text prompt (or reference art) build a physically real world, no splats, no diffusion, but full physics? Once you have this, can a robot or biocomputer navigate it? Full demo of the system running live (local / remote hybrid) How it is all plumbed and built. The system uses a variety of learning strategies including self-modifying code and rules to both iteratively improve a world model and, iteratively improve the generation of that model AND, iteratively improve the improvement of that model (3 level hierarchy). It uses a large range of optimising strategies to avoid uncanny valley effects and attempt to produce cinematically realistic images.

- Event context: AI Tinkerers San Diego: February Meetup at Google — 2026-02-27 — San Diego
- Public talk page: https://san-diego.aitinkerers.org/talks/rsvp_mnJaK0Bd4l8

### [Poor Man’s Long Running Multi Hour Coding Agent](https://seattle.aitinkerers.org/talks/rsvp_2_7aWPHU1uc)

In December 2025, I was running out of time. In an act of desperation and after reading Anthropix's blog post or paper about long-running agentic harnesses, I attempted to replicate what they described in their paper and was able to achieve over a long weekend the completion of a feature done enough and well enough that I could ship and meet my deadline. The outcome of this was over that weekend I completely maxed out my Claude Max subscription and then on top of that I incurred an overage of $700.

- Event context: AI Tinkerers Seattle Meetup: Dev Tools Track — February 9th, 2026 — 2026-02-10 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_2_7aWPHU1uc

### [Hooks are Not Enough: Securing the Network Layer for Agents](https://sf.aitinkerers.org/talks/rsvp_gFL3GVRbRq8)

I'll be presenting hands-on approaches to implementing network access controls for agents, specifically Claude Code. I'll start with a simple built-in solution, show why that's insufficient, and work up to a more secure solution.

- Event context: AI Tinkerers x Google Cloud SF Meetup — 2025-11-21 — San Francisco
- Public talk page: https://sf.aitinkerers.org/talks/rsvp_gFL3GVRbRq8

### [De la pereza a la automatización](https://manizales.aitinkerers.org/talks/rsvp_rEVE7dOTXik)

En esta charla voy a contar mi experiencia a lo largo de los años con diferentes formas de automatizar, proyectos y tareas del día a día

- Event context: 🏔️ ¡Noveno Encuentro de AI Tinkerers Manizales! — 2025-10-29 — Manizales
- Public talk page: https://manizales.aitinkerers.org/talks/rsvp_rEVE7dOTXik

### [Simple Retrieval with Codex CLI and Bash](https://seattle.aitinkerers.org/talks/rsvp_9-gmI6Gpuh0)

Parallel document extraction demo using multiple Codex CLI instances running simultaneously. Each agent uses the experimental_instructions_file field to receive specialized prompts for extracting different information (payment amounts, deadlines, parties, etc.) from contracts; no embedding pass needed. Live bash script demo with timestamped results.

- Event context: October Meetup - Science Fair at Foundations — 2025-10-23 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_9-gmI6Gpuh0

### [Local Transcription](https://valencia.aitinkerers.org/talks/rsvp_VfuECFTODQ8)

Too many meetings and recordings have private information that you dont want shared with 3rd party transcription services and platforms. I built a simple terminal app that transcribes all my meetings and recordings without using the platform automations. This runs with one command and does file cleanup. Its simple and rough, but it gets the job done reliably.

- Event context: Opening AI Tinkerers Valencia x VDS - 22 October — 2025-10-22 — Valencia
- Public talk page: https://valencia.aitinkerers.org/talks/rsvp_VfuECFTODQ8

### [Orquestación y especialización de agentes código](https://medellin.aitinkerers.org/talks/rsvp_53m46l-k0-k)

Un camino en como de manera personal orquestro reviso y especializo agentes para ser 10x developer.

- Event context: AI Tinkerers Medellín #16 - Lean Solutions Group - 25 de Septiembre — 2025-09-26 — Medellín
- Public talk page: https://medellin.aitinkerers.org/talks/rsvp_53m46l-k0-k

### [AI Task Hub](https://nurnberg.aitinkerers.org/talks/rsvp_sevtdmGHToE)

What I'm currently implementing and will demonstrate could be described as similar to PromptHub (https://www.prompthub.us/). The key difference is that it's more user and developer-participatory. While PromptHub focuses primarily on prompts, my solution allows developers to define AI tools in more flexible units, enabling sharing and team-based development.

- Event context: AI Tinkerers Nürnberg First Meetup - July 3rd — 2025-07-03 — Nürnberg
- Public talk page: https://nurnberg.aitinkerers.org/talks/rsvp_sevtdmGHToE

### [You don't need the Internet to use an LLM](https://nashville.aitinkerers.org/talks/rsvp_-iX7E_vYXM4)

This demo shows multiple ways to leverage the power of LLMs without interacting with remote web services. The models will be run in a local and portable manner, covering how to tweak the parameters used to dictate how various models run in resource-constrained environments.

- Event context: AI Tinkerers Nashville – June 23rd, 2025 — 2025-06-23 — Nashville
- Public talk page: https://nashville.aitinkerers.org/talks/rsvp_-iX7E_vYXM4

### [Deploying fine-tuned language models: From start to finish, in nothing but Python](https://toronto.aitinkerers.org/talks/rsvp_6a2kJuTUIXQ)

What does it take to fine-tune and deploy a customized LLM on state-of-the-art cloud hardware? In this talk, we explore a fully Pythonic solution to this problem, using just a few extra lines on top of ordinary code. No cloud expertise is required to follow along. We’ll start with a simple example and scale things gradually to arrive at a powerful, high-compute workflow that creates a “model inference” service with a custom API—all within the confines of a Jupyter Notebook!

- Event context: AI Tinkerers - Toronto - May 2024 Meetup — 2024-05-30 — Toronto
- Public talk page: https://toronto.aitinkerers.org/talks/rsvp_6a2kJuTUIXQ

## Related Technologies

- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 654 public demos
- [Claude](https://aitinkerers.org/technologies/claude) ([Markdown](https://aitinkerers.org/technologies/claude.md)) — 170 public demos
- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 204 public demos
- [JavaScript](https://aitinkerers.org/technologies/javascript) ([Markdown](https://aitinkerers.org/technologies/javascript.md)) — 40 public demos
- [TypeScript](https://aitinkerers.org/technologies/typescript) ([Markdown](https://aitinkerers.org/technologies/typescript.md)) — 198 public demos
- [Anthropic API](https://aitinkerers.org/technologies/anthropic-api) ([Markdown](https://aitinkerers.org/technologies/anthropic-api.md)) — 63 public demos
- [BNF Grammar](https://aitinkerers.org/technologies/bnf-grammar) ([Markdown](https://aitinkerers.org/technologies/bnf-grammar.md)) — 1 public demo
- [ChatGPT](https://aitinkerers.org/technologies/chatgpt) ([Markdown](https://aitinkerers.org/technologies/chatgpt.md)) — 83 public demos
- [Code](https://aitinkerers.org/technologies/code) ([Markdown](https://aitinkerers.org/technologies/code.md)) — 9 public demos
- [Codex](https://aitinkerers.org/technologies/codex) ([Markdown](https://aitinkerers.org/technologies/codex.md)) — 43 public demos
- [Codex CLI](https://aitinkerers.org/technologies/codex-cli) ([Markdown](https://aitinkerers.org/technologies/codex-cli.md)) — 5 public demos
- [Covalent](https://aitinkerers.org/technologies/covalent) ([Markdown](https://aitinkerers.org/technologies/covalent.md)) — 1 public demo
- [Drupal](https://aitinkerers.org/technologies/drupal) ([Markdown](https://aitinkerers.org/technologies/drupal.md)) — 1 public demo
- [FastAPI](https://aitinkerers.org/technologies/fastapi) ([Markdown](https://aitinkerers.org/technologies/fastapi.md)) — 178 public demos
- [FFmpeg](https://aitinkerers.org/technologies/ffmpeg) ([Markdown](https://aitinkerers.org/technologies/ffmpeg.md)) — 15 public demos
- [git](https://aitinkerers.org/technologies/git) ([Markdown](https://aitinkerers.org/technologies/git.md)) — 23 public demos
- [git grep](https://aitinkerers.org/technologies/git-grep) ([Markdown](https://aitinkerers.org/technologies/git-grep.md)) — 1 public demo
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 73 public demos