Forge

by Carlo Perottino

Things I need. Things I want. Things that interest me.

Exploring Agent Auth
Tempered

Exploring Agent Auth

Eight identity & authorization patterns for AI agents calling tools

Python Jupyter Docker
+4
Coming Soon

LLM vs. The Right Tool

Head-to-head notebooks pitting LLMs against the classical methods built for the job

Python LLM ML
+1
Coming Soon

LLM Playground

Run the same prompt against flagship models side-by-side and see exactly where they differ

TypeScript LLM React
Coming Soon

ISP Watchdog

Continuous internet speed monitoring with anomaly detection and automated provider accountability

Python AWS Raspberry Pi
+1
Coming Soon

Social AI Lab

A framework for running socialization experiments between LLMs — games, debates, alignment tests

Python LLM Benchmarking

Elsewhere on GitHub

Standalone projects and experiments

A few things I’ve made outside Forge—useful, interesting, and still worth sharing.

GitHub project

NetNudge

AI-powered contact management and personalized outreach for your professional network.

Python AI Agents Streamlit Google Contacts
GitHub project

Likeness Lab

A practical AWS workflow for training a personal likeness LoRA from a folder of photos.

Python LoRA Flux AWS
GitHub project

FutureBucks

A life-event-driven financial simulation tool for long-range scenarios and projections.

Python Finance Simulation Streamlit
GitHub project

Transformer From Scratch

A weekend-sized, hands-on path to building and training a complete GPT from raw PyTorch operations.

Python LLMs PyTorch Education
GitHub project

Perspectv

AI-powered competitive intelligence engine for generating personalized executive outreach messages in B2B sales.

Python AI Agents RAG B2B Sales
GitHub project

Prism

A gallery and workbench that runs one prompt across flagship models, aligns their outputs, and surfaces meaningful differences.

TypeScript LLM Comparison Output Diffing BYO Keys
GitHub project

LLM Lifecycle / Crucible

An interactive visualization of the full LLM lifecycle, replaying real training runs from initialization through pre-training, SFT, reinforcement learning, and inference.

Python LLM Training Reinforcement Learning Visualization
GitHub project

SkyTunnel

A self-hosted toolkit that deploys DNS, ICMP, and HTTPS tunnels on EC2 for resilient connectivity in restricted networks.

Shell DNS / ICMP / HTTPS AWS CloudFormation SOCKS5
GitHub project

OpenAI Cookbook Sandbox

Ordered, runnable exercises for studying the OpenAI API stack, from Responses and built-in tools to retrieval, agents, and production patterns.

Jupyter Notebook OpenAI API Responses API Agents SDK
GitHub project

Text2SQL Perf Bench

A local benchmark harness for measuring generated SQL's end-to-end runtime across query-complexity levels and context strategies, using OpenAI models for generation.

SQL Benchmarking OpenAI