GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles
  • Enterprises

This view

  • Original page
  • New address

GeezerVison is showing github.com — samestrin - Overview

@samestrin samestrin Follow
More
  • Overview
  • Repositories
  • Projects
  • Packages
  • Stars
@samestrin
samestrin
Follow
View samestrin's full-sized avatar

Sam Estrin samestrin

Follow
Principal Engineer @ Estrin Group. Deterministic First AI: agents, evaluation, production AI. 8x CTO. Co-founder of 2 public companies.
19 followers · 0 following
  • San Diego, California
  • https://samestrin.com

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx4Achievement: StarstruckAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx4Achievement: StarstruckAchievement: YOLOAchievement: Quickdraw
More
  • Overview
  • Repositories
  • Projects
  • Packages
  • Stars
samestrin/README.md

Sam Estrin

Principal Engineer at Estrin Group. I build products and AI systems for clients and serve as a fractional CTO. Eight times CTO, including CTO of the Evidence.com division at TASER (now Axon). Co-founder of 2 companies that went public.

I design AI systems on a Deterministic First principle. Deterministic code owns everything it can do reliably. LLMs are limited to the steps that require judgment. Every model output is validated in code before the workflow continues, with human approval for critical actions.

Start here:

  • atcr: Agent Team Code Review. A panel of AI reviewers on different models, merged by a deterministic reconciler. Go.
  • llm-interface: one interface for 36 LLM providers. JavaScript, npm.
  • llm-tools: MCP servers and CLIs for AI agents. Go.
  • llm-env: point any OpenAI- or Anthropic-compatible tool at any provider. Bash.

samestrin.com · estringroup.co · LinkedIn

Pinned Loading

  1. llm-interface llm-interface Public

    A simple NPM interface for seamlessly interacting with 36 Large Language Model (LLM) providers, including OpenAI, Anthropic, Google Gemini, Cohere, Hugging Face Inference, NVIDIA AI, Mistral AI, AI…

    JavaScript 123 13

  2. llm-env llm-env Public

    Force any OpenAI-compatible or Anthropic API-compatible tool (Claude Code, Qwen Code, Aider, Fabric, Interpreter) to use Gemini, Groq, Cerebras, or Ollama. Pure Bash. Zero dependencies.

    Shell 9 1

  3. llm-pdf-ocr-api llm-pdf-ocr-api Public

    A Python-based REST API for PDF OCR using AI models with PyTorch and Transformers that runs in a Docker container.

    Python 37 11

  4. llm-newsletter-generator llm-newsletter-generator Public

    llm-newsletter-generator transforms a valid RSS feed into a "Newsletter" using AI models via PyTorch and Transformers; this is experimental.

    Python 15 4

  5. llm-tools llm-tools Public

    High-performance Go tools for LLM Agents (MCP/CLI). 100x faster than Python. Includes native MCP Server, token-optimized search, and persistent project memory.

    Go 3 1

  6. atcr atcr Public

    Agent Team Code Review

    Go

Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.

Uh oh!

There was an error while loading. Please reload this page.

You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.