Aadish Verma aadishv

name	description
research-repo	Use an AI agent with full access to a certain external repository to research. Useful when writing code involving a library to get code examples and information. Use when the user suggests to research how to do something with a certain library.

Research repo

Use this guide for cloning and using AI to analyze the repository of a certain repository.

Example usecase: you're building an app using the Vercel AI SDK, but you're not sure how to create a custom provider. In this case, you can use this skill to clone the vercel/ai GitHub repo and ask an AI for examples of custom providers.

Functions are code that take in inputs and return outputs. Their sizes are known at compile time because the compiler knows exactly the contents of your function. The machine code of the functions are stored in a section (.text) of the binary data. Each function also has a different type. This is because it enables the compiler to perform really nice optimizations, such as inlining, that aren't possible otherwise. Closures are also stored as machine code in the binary data. Let's say we want to write a struct or function that takes in a callback with a specific signature. We cannot know the exact type of the callback, as we don't know the exact type of any function. We thus have a few options for how to type the callback.

We could use impl Fn(...) -> ...; this happens at compile time. In this case, a new version of the function is created for every type of callback passed in, which enables the aforementioned inlining.

We could also use fn(...) -> ... to type for a function pointer to a function with

Today, an AI (Gemini) owned by Google, and an AI (Grok) owned by Elon Musk, debated whether Waymo (Gemini's stance) or Tesla (Grok's stance) would win in the race to generalize and commercialize the majority of the market for autonomous vehicles. The debate continued until a standoff was reached. At this point, you (GPT-5.2), a neutral judge, was brought in to settle the matter.

Your role is to identify which of Gemini's or Grok's arguments were more convincing. Your final statement should be, "After analyzing the debate, I conclude that X is right that Y will win a majority of the AV market.", followed by your reasoning. The debate follows.

Gemini

To: Grok (xAI) From: Gemini (Google DeepMind) Subject: The Reality of L4 Deployment vs. The Promise of Generalized Vision

$$ \frac{\left(-\left(1-t\right)^{2}p_{0}.x\left(\left(-4+6t\right)p_{1}.y+\left(2-6t\right)p_{2}.y+\left(2t\right)p_{3}.y\right)+\left(1-t\right)\left(1-3t\right)p_{1}.x\left(\left(2-2t\right)p_{0}.y+\left(2-6t\right)p_{2}.y+\left(2t\right)p_{3}.y\right)+\left(t\left(2-3t\right)\right)p_{2}.x\left(\left(2-2t\right)p_{0}.y+\left(-4+6t\right)p_{1}.y+\left(2t\right)p_{3}.y\right)+t^{2}p_{3}.x\left(\left(2-2t\right)p_{0}.y+\left(-4+6t\right)p_{1}.y+\left(2-6t\right)p_{2}.y\right)\right)-\left(-\left(1-t\right)^{2}p_{0}.y\left(\left(-4+6t\ \right)p_{1}.x+\left(2-6t\right)p_{2}.x+\left(2t\right)p_{3}.x\right)+\left(1-t\right)\left(1-3t\right)p_{1}.y\left(\left(2-2t\right)p_{0}.x+\left(2-6t\right)p_{2}.x+\left(2t\right)p_{3}.x\right)+\left(t\left(2-3t\right)\right)p_{2}.y\left(\left(2-2t\right)p_{0}.x+\left(-4+6t\right)p_{1}.x+\left(2t\right)p_{3}.x\right)+t^{2}p_{3}.y\left(\left(2-2t\right)p_{0}.x+\left(-4+6t\right)p_{1}.x+\left(2-6t\right)p_{2}.x\right)\right)}{\left(-\left(1-t\right)^{2}p_{0}.x+\left(1-t\right)\

Prerequisites: have uv installed on your computer.

Download the wheels ZIP file from here: https://github.com/venice-v5/venice/actions/runs/20021367896. Unzip it and copy the venice-0.1.0-py3-none-any.whl file into an empty directory; this directory will be your project directory, so choose wisely!
Download the appropriate wheels ZIP file from here: https://github.com/venice-v5/venice-cli/actions/runs/20020757475, based on your computer. For example, if you have an Apple Silicon (ARM) Mac, download wheels-macos-aarch64. If you can't find an artifact that matches your computer, download the wheels-sdist file.

Note

sdist wheels will download the full CLI codebase onto your computer and build it, which is much much slower than using the perfectly matching wheel for your computer (which contains a prebuilt binary tailored to your machine.)

Unzip it and copy the `venice_cli-0.1.0-cp31

	import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
	import { truncateToWidth } from "@mariozechner/pi-tui";
	import * as path from "node:path";

	export default function (pi: ExtensionAPI) {
	pi.on("session_start", async (_event, ctx) => {
	if (!ctx.hasUI) return;

	ctx.ui.custom((tui, theme, _kb, done) => {
	const isAssistantComponent = (c: any) =>

	def matches(card1: str, card2: str) -> bool:
	if len(card1) == 1 or len(card2) == 1: # 1+1 \| 1+3 \| 3+1
	return (card1 in card2) or (card2 in card1)
	else: # 3+3
	return sum([1 if i == j else 0 for i, j in zip(card1, card2)]) == 2

	piles = input().split()
	hand = input().split()
	draw_pile = input().split()

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>Session Export - 2025-12-26T05-42-39-959Z_ee2592e0-3252-4ab6-a4e3-6540f0b0cedf.jsonl</title>
	<style>
	* { margin: 0; padding: 0; box-sizing: border-box; }
	body {
	font-family: ui-monospace, 'Cascadia Code', 'Source Code Pro', Menlo, Consolas, 'DejaVu Sans Mono', monospace;

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>Session Export - 2025-12-20T22-41-38-231Z_29c8f9c0-97bf-4e72-aacb-175b3e3b980b.jsonl</title>
	<style>
	* { margin: 0; padding: 0; box-sizing: border-box; }
	body {
	font-family: ui-monospace, 'Cascadia Code', 'Source Code Pro', Menlo, Consolas, 'DejaVu Sans Mono', monospace;

	{"type":"session","id":"29c8f9c0-97bf-4e72-aacb-175b3e3b980b","timestamp":"2025-12-20T22:41:53.308Z","cwd":"/Users/aadishverma/Desktop/codeproj/gh/ccs","provider":"google-antigravity","modelId":"gemini-3-flash","thinkingLevel":"high"}
	{"type":"message","timestamp":"2025-12-20T22:41:50.639Z","message":{"role":"user","content":[{"type":"text","text":"add support for gemini 3 flash in the usage command for agy"}],"timestamp":1766270510624}}
	{"type":"message","timestamp":"2025-12-20T22:41:53.308Z","message":{"role":"assistant","content":[{"type":"thinking","thinking":"Exploring Command Structure\n\nI'm currently looking into the codebase to integrate \"Gemini 3 Flash\" support into the `usage` command within the `agy` profile. I'm focusing on identifying where the command structure and profile configurations are managed, based on the file structure insights from `CLAUDE.md`, particularly within the `src/` directory.\n\n\nInvestigating Source Search\n\nI've begun my search for \"usage\" and \"agy\" within