50 filings on the bench

The Lab Bench

Empirical explorations, reproducible recipes, interactive benchmarks, and field monographs — published as experiments land. Proven patterns graduate downstream into Standards ↗ and Courses ↗.

Format
Topic

Showing 50 of 50 filings

№9 Soft Clay · Q4 2026
EXPERIMENT The Harness

Adapting Qwen 3.8 27B to the π Coding Harness on Apple Silicon

Testing harness-hub's RL LoRA for the π agent on a 64 GB M4 Max: why Q4_K_M hits 19.45 tok/s (+34% over Q8) while Q3_K_M regresses on Metal, how to keep .pi project-local with mise, and timing agent loops from llama.cpp Prometheus counters.

EXPLORING
6 min
№9 Soft Clay · Q4 2026
RECIPE Surfaces & UX

Repeatable Site Demo Reels with Claude Code, webreel, and Remotion

How we evolved a one-off webreel + Python site trailer into a repeatable Remotion 4 studio — with swappable TheFocus.AI bumpers, an interactive bumper picker, a drop-in reel.json format, and one-command cuts for Slack, X, and YouTube.

12 min
Bound Edition · Q4 2026
STATE № 03 · MONOGRAPH Field & State

STATE № 03: The San Francisco Report

Software factories force a complete rethink of the SDLC — changing how features get specified going in, and how frozen verification gates what comes out.

VERIFIED
16 min
№9 Soft Clay · Q4 2026
BENCHMARK · RM-2026-11-GEMMA2 Models & Local

Full Multimodal Search Without the Data Leaving the Building

EmbeddingGemma 2 + one SQLite file · 16 ms · zero egress. Ingest PDFs, PowerPoints, scans, code, and audio (~2 bimodal pages/s), search them locally, and read an honest tool scorecard.

VERIFIED → STANDARD → COURSE
8 min
№9 Soft Clay · Q4 2026
BENCHMARK · RN-2026-11-NQPF Org Age

New productive forces. New organizations?

An interactive research notebook on 新质生产力: change a factory, explore the terminology, and test how AI capabilities might change roles, coordination, and authority.

VERIFIED → STANDARD → COURSE
8 min
Bound Edition · Q4 2026
STATE № 02 · MONOGRAPH Field & State

STATE № 02: The New York Report

Half of large enterprises run coding agents, yet only five percent have autonomous agents working anywhere else in the business — crossing that 95% diffusion wall takes bespoke evals, runtime policy governance, and real tokenomics.

VERIFIED
16 min
№8 Crosstalk · Q3 2026
BENCHMARK · RM-2026-09-JEV The Harness

Jev: Judging, Not Calling Tools

A new LLM interface: instead of a continuation, a tool call, or a free-form completion, you ask a fixed set of judgments — a high-speed classifier. Umwelten’s types, a claim-and-passage walkthrough, and a Jev vs DeepSeek pilot.

GRADUATED → STANDARD → COURSE
8 min
№8 Crosstalk · Q3 2026
BENCHMARK · FR-2026-07-PHONE The Harness

Never Touch the Laptop: Running a Whole Project from Your Phone

Grok Bot dispatching to Cursor Cloud Agents, Cursor environment setup, Amp orbs, Codex Cloud, and Claude Code on the web mapped across 13 axes — voice, secrets, MCP, previews, and the build loop — against the idealized phone-only project.

GRADUATED → STANDARD
8 min
№8 Crosstalk · Q3 2026
EXPERIMENT Models & Local

whichllm: pick a local model that is actually fast

whichllm ranks the best local models for your hardware — general, coding, vision, math — then whichllm run downloads the top pick and starts a chat. --speed fast keeps only the ones estimated at 30+ tok/s.

GRADUATED → COURSE
5 min
№8 Crosstalk · Q3 2026
BENCHMARK · RM-2026-07-KV Models & Local

Conversation Roots, KV Reuse, and Local Agent Throughput

How to win local KV reuse: exact prefix rules, slot sizing, and measured results across Ollama and llama.cpp on a 64 GB M4 Max—same mechanism, one study.

VERIFIED → COURSE
8 min
№8 Crosstalk · Q3 2026
BENCHMARK · FR-2026-07-AIEWF Field & State

AI Engineer World's Fair 2026 — The Complete Talk Roundup

Every talk from the fair, read and distilled — 92 write-ups from ~319K words of transcripts, nine topic tracks, and the cross-cutting surprises.

VERIFIED
8 min
№7 Dispatches · Q2 2026
BENCHMARK · RM-2026-05-LP Models & Local

Local Providers: Which Open Model Should You Actually Run?

We tested 7 model families across 3 local runtimes on a 64 GB Mac — 20 cells, 5 dimensions.

VERIFIED
8 min
Bound Edition · Q2 2026
STATE № 01 · MONOGRAPH Field & State

STATE № 01: The Miami Report

The difficulty of software engineering used to act as a hidden filter against bad ideas — and we just removed it.

VERIFIED
16 min
№6 Proofs · Q1 2026
BENCHMARK · RM-2026-03-MS Models & Local

Testing 48 LLMs Across 5 Dimensions for $4.63

Interactive leaderboard across reasoning, knowledge, instruction following, coding, and MCP tool use.

VERIFIED
8 min
№6 Proofs · Q1 2026
BENCHMARK · RM-2026-03-MCP The Harness

Can LLMs Use Real-World Tools?

39 models, real Rivian driving data via MCP — most can call tools; fewer call the right ones.

VERIFIED
8 min
№6 Proofs · Q1 2026
BENCHMARK · RM-2026-02-CW Models & Local

The Car Wash Test: Do LLMs Have Common Sense?

131 models, one question — should you walk or drive to the car wash?

VERIFIED
8 min
№5 Instruments · Q4 2025
RECIPE The Harness

Build Your Own Coding Agent

A step-by-step tutorial for building an AI coding assistant from scratch. Bootstrap it with an existing agent, then use it to improve itself.

1 min
№5 Instruments · Q4 2025
RECIPE The Harness

The Smallest Coding Agent

A complete agentic coding loop in 25 lines of Bun. No frameworks, no dependencies, just bash and a model.

1 min
№5 Instruments · Q4 2025
BENCHMARK · RM-2025-11-AIECODE Field & State

AI Engineering Code Summit 2025

Summit report — talks, sessions, and takeaways from Miami.

VERIFIED
8 min
№4 The Console · Q3 2025
RECIPE The Harness

Using repomix to give llm full context

How to use repomix to package your entire codebase into a single file for LLM context, with tips for managing token counts and mise integration.

1 min
№4 The Console · Q3 2025
RECIPE Field & State

context7-mcp

Learn how to integrate Context7 MCP server to get up-to-date code documentation for LLMs and AI code editors, enabling real-time access to the latest library and framework documentation.

1 min
№4 The Console · Q3 2025
RECIPE Field & State

use playwright mcp to get high fidelity designs

Learn how to integrate Playwright MCP server to enable AI coding agents to interact with web browsers, take screenshots, and validate designs with high fidelity for accurate web development and testing.

1 min
№4 The Console · Q3 2025
EXPERIMENT The Harness

Single file swift mini-apps

swift files can be run directly without compiling and without XCode, making it easy to create native UI elements and access all of macOS's APIs. Once you see Swift as a scripting language rather than just an app language, you start wondering what other capabilities are hiding in plain sight.

3 min
№4 The Console · Q3 2025
EXPERIMENT Models & Local

Code Generation with Local Models

Small, local AI models deliver surprisingly effective results for everyday tasks. Also llama3.2 is surprisingly fast and gpt-oss is surprisingly good.

2 min
№4 The Console · Q3 2025
RECIPE The Harness

Dagger and Container Use

Unlock the power of isolated development with Dagger and Container Use—no more worrying about missing dependencies or system conflicts.

1 min
№3 Field Trials · Q2 2025
RECIPE Surfaces & UX

Use Cursor in Slack

Learn how to integrate Cursor with Slack for seamless code review and collaboration, including setup tips and workflow enhancements.

1 min
№3 Field Trials · Q2 2025
BENCHMARK · RM-2025-06-CAR The Harness

June 2025 Coding Agent Report

15 AI coding tools tested — strengths, weaknesses, and recommendations.

VERIFIED
8 min
№3 Field Trials · Q2 2025
RECIPE Surfaces & UX

Content Finder

Ever had a movie stuck on the tip of your tongue or wanted more like your family's favorite? Explore how AI-powered content discovery can help you make your next pick.

1 min
№3 Field Trials · Q2 2025
RECIPE Field & State

Geolocation Prompt

This post introduces a prompt for high-precision, image-based geolocation analysis, guiding AI to deduce where a photo was taken using expertise in visual, cultural, and environmental cues. The prompt provides a step-by-step framework for analyzing images and generating confident, reasoned location guesses—even suggesting what extra information would help if uncertain.

1 min
№3 Field Trials · Q2 2025
RECIPE Surfaces & UX

Iterating designs with Chat GPT 4o

Discover how to leverage Chrome's hidden command palette and ChatGPT to transform your website screenshots into sophisticated wireframes. Learn a clever workflow that combines Chrome's full-page screenshot capability with ChatGPT's AI-powered design suggestions to create professional 16:9 ratio mockups for your next project.

1 min
№3 Field Trials · Q2 2025
RECIPE Surfaces & UX

Use the llm cli for quick prototyping

Transform your workflow with a command-line tool that brings AI capabilities directly to your terminal, no complex setup required. See how to parse websites, analyze transcripts, and extract meaningful data using the llm CLI with popular models like Gemini and Claude.

2 min
№3 Field Trials · Q2 2025
RECIPE The Harness

Lift and shift

Master the art of learning from existing projects by using this structured approach to extracting and documenting requirements and insights. With simple command-line tools and clear documentation practices, you'll be able to turn any existing implementation into a wealth of reusable knowledge.

1 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Claude to write midjourney prompts

Transform your written content into compelling visual concepts by using Claude as an intelligent intermediary for Midjourney prompt creation. See how combining prompt libraries with AI analysis can help you explore multiple creative directions before settling on the perfect visual representation.

1 min
№2 Blueprints · Q1 2025
EXPERIMENT Surfaces & UX

Schema-Driven AI: Better User Experiences with Structured Output

Transforms chatting from simple text generators into powerful data processing engines, enabling extraction of organized information from PDFs, audio files, and more. Here are some practical techniques for building, including audio analysis, pdf data extraction and conversation state management, showcasing how constraint-driven outputs can power rich user experiences.

12 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Make it run tests

1 min
№2 Blueprints · Q1 2025
EXPERIMENT The Harness

Image Gen on Apple Silicon

We've got the apple silicon, lets download some models and make some pictures

1 min
№2 Blueprints · Q1 2025
RECIPE Surfaces & UX

Adding Memory to Cursor

Transform your Cursor IDE into a more intelligent assistant by implementing a sophisticated memory system adapted from cline-memory-bank. This enhanced version includes TDD principles and integrates seamlessly with your prompt library, allowing your AI assistant to maintain context and remember important details across multiple coding sessions.

1 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Planning, brainstorming and idea development

Transform your rough concepts into detailed project specifications with this powerful prompting technique that guides you through an in-depth development process. By asking just one thoughtful question at a time, AI helps you create comprehensive, developer-ready plans that cover every essential detail.

1 min
№2 Blueprints · Q1 2025
RECIPE Surfaces & UX

Make it easy to load and update your prompt library

Never lose track of your valuable AI prompts again with this elegant solution for maintaining a personal prompt library. The system provides simple commands to list, load, and save prompts, making it effortless to reuse and iterate on your most effective conversation starters.

1 min
№2 Blueprints · Q1 2025
RECIPE Surfaces & UX

Superwhisper for input

Imagine talking to your computer and having it type for you—Superwhisper makes it real. Check out how this tool speeds up input and even lets you tweak the output.

1 min
№2 Blueprints · Q1 2025
RECIPE Surfaces & UX

Talking to the models in the car

Discover how voice-activated assistants like Google and ChatGPT can transform your car rides into interactive conversations. Learn tips like using search and prompts to dive deeper into news or ideas while on the go.

2 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Extracting requirements

Need to understand all the requirements hidden in a codebase? This technique shows how to use LLMs to compile comprehensive specifications from repository analysis for immediate development use.

1 min
№2 Blueprints · Q1 2025
EXPERIMENT The Harness

Exposing Services with MCP

Model Context Protocol bridges the gap between AI models and your applications. Learn how defining simple tools with descriptions and parameters lets Claude intelligently combine services without explicit instructions.

3 min
№2 Blueprints · Q1 2025
RECIPE The Harness

Write git-commit-messages

Learn how to automate Git commit message generation with this simple but powerful prompt for Claude. The template enforces conventional commit format with appropriate type prefixes and detailed descriptions, making your repository history more professional and useful.

1 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Making Meta Data

Generating metadata doesn't have to be a manual process when you can leverage AI prompts to produce multiple options quickly. Discover how to implement command-line tools that streamline metadata creation while maintaining your role as the final decision-maker.

1 min
№2 Blueprints · Q1 2025
RECIPE Surfaces & UX

Fetch and Brave Search in Claude Desktop

1 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Copy a podcast into notebookllm

1 min
№2 Blueprints · Q1 2025
RECIPE The Harness

Using repomix to give llm full context (original)

The post focuses on how to give LLMs full visibility into a codebase to improve their understanding and responses.

1 min
№2 Blueprints · Q1 2025
RECIPE Field & State

Understanding a code base

Explore three complementary approaches to rapidly comprehend any codebase using the latest AI assistants and developer tools. Follow along with practical examples showing how to extract key information about memory systems and other complex functionality with minimal effort.

1 min
№1 Sketches · Q4 2024
EXPERIMENT Surfaces & UX

Slicing up a design from figma

In this hands-on comparison, three coding tools - Cursor, Aider, and v0 - are put to the test as they attempt to replicate a design from Figma into functional HTML and CSS code, revealing their strengths, weaknesses, and quirks.

5 min
Loading archive index…