Vivek Haldar vivekhaldar

Software Design Document

1. Introduction

This document describes a command-line Python application for Automatic Time Tracking by Watching Computer Screen. The application periodically captures screenshots on macOS, processes them using a vision-based language model, and generates time-tracking reports.

The design covers:

Overall architecture
System components
Data flows
Implementation details

Product Specification: Automatic Time Tracking by Watching Computer Screen

1. Overview

Goal

The objective of this software is to automatically measure and analyze how time is spent on a computer by periodically capturing screenshots, using a vision-based language model (LLM) to describe the content, and summarizing the results to generate a high-level report of time allocation across applications and tasks.

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>TheAgentCompany: Deep Dive</title>
	<link rel="preconnect" href="https://fonts.googleapis.com">
	<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
	<link href="https://fonts.googleapis.com/css2?family=Space+Grotesk:wght@400;500;600;700&family=Inter:wght@300;400;500;600&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">
	<style>

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>AI Agents in Semiconductor Test Analytics \| Market Survey 2025</title>
	<link rel="preconnect" href="https://fonts.googleapis.com">
	<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
	<link href="https://fonts.googleapis.com/css2?family=Instrument+Serif:ital@0;1&family=JetBrains+Mono:wght@400;500&family=Plus+Jakarta+Sans:wght@300;400;500;600;700&display=swap" rel="stylesheet">
	<style>

	{
	"sections": [
	{
	"title": "Introduction and Utility of ChatGPT",
	"summary": [
	"The use of ChatGPT has significantly impacted the guest's personal and professional life, offering assistance in learning languages, understanding complex topics, and enhancing travel experiences.",
	"ChatGPT can act as a universal translator, providing assistance with menu selection in foreign countries and helping identify unknown plants or animals.",
	"The tool is seen as an integral part of life, deeply intertwined with the user's ability to work efficiently and learn new information."
	],
	"quotes": [

	{
	"sections": [
	{
	"title": "Tyler Cowen's use of ChatGPT",
	"summary": [
	"Uses ChatGPT on iPhone as a universal translator when traveling in foreign countries",
	"Uses it to read menus, identify plants and animals, and get recommendations",
	"On laptop, uses it to learn about obscure history and get background context for interviews",
	"Asks follow-up questions to probe deeper and fact-check potential hallucinations"
	],

	import os
	from time import sleep
	from openai import OpenAI

	client = OpenAI(
	# defaults to
	api_key=os.environ.get("OPENAI_API_KEY"),
	)

	# Step 1: Create an Assistant

	;; emacs-gpt.el -- Control Emacs via ChatGPT
	;;
	;; Basic idea: take natural language input from user, ask ChatGPT for
	;; corresponding elisp, run it.


	(defun mark-between-assistant-and-user ()
	"Mark the region between \"%assistant%\" and \"%user%\", not including those strings."
	(interactive)
	(goto-char (point-min))

	#!/usr/bin/env python3
	#
	# Takes a chat transcript (for ChatGPT) on stdin, calls the OpenAI
	# ChatGPT API, and prints the response on stdout.
	#
	# Your OpenAI API key must be set in the environment variable
	# OPENAI_API_KEY.
	#
	# Logs are written to ~/chat.log.