1 Million Lines of AI-Assisted Code in 2025

How AI transformed my coding productivity from 24k lines in 2023 to over 1 million in 2025 — a 40x increase while only coding 1-2 days per week

By Jonathan ClarkFebruary 1, 202510 min read
← Back to Blog

Using AI, I will have "written" over 1 million lines of code this year, compared to 57k in 2024 and 24k in 2023. The chart below shows how much cumulative code I've written since 2023, and you can see progress accelerating — yet I only do focused coding 1-2 days per week.

1M+
Lines in 2025
↑ 1,754% from 2024
260+
Total Projects
Since 2023
1-2
Days Per Week
Focused Coding
2M+
Expected 2026
↑ 100% projection

Cumulative Lines of Code Over Time

The AI Acceleration Effect

The exponential growth you see in the chart isn't just about writing more code — it's about a fundamental shift in how software gets built. AI has transformed coding from typing-limited to idea-limited. The bottleneck is no longer implementation but imagination.

What's remarkable is that this acceleration happened while maintaining a part-time coding schedule. I typically dedicate only 1-2 days per week to focused development, yet the output has increased by 40x in just two years.

Breaking Down the Numbers

Yearly Comparison

Technology Stack Distribution

The diversity of technologies used across these 260+ projects reflects the breadth of problems AI can help solve. From web applications to data analysis, from automation scripts to machine learning models, each language serves its purpose in the ecosystem.

Lines of Code by Language/Technology

JavaScript (296k lines) and HTML (278k lines) dominate the codebase, reflecting a heavy focus on web development and browser-based applications. Python (87k lines) comes in third, powering data analysis, automation, and AI integration projects. The presence of C/C++ (48k lines) shows involvement in performance-critical and system-level programming.

The data tells a clear story:

  • 2023: 24,200 lines across 61 projects — The baseline year, mostly manual coding
  • 2024: 57,795 lines across 106 projects — Early AI adoption, 2.4x increase
  • 2025: 810,605+ lines across 108 projects (and counting) — Full AI integration, 14x increase from 2024

Beyond Code: Document Engineering & Legal Tech

While the million lines metric captures code output, AI has equally transformed my document and legal work. A significant portion of my projects involve sophisticated document generation, processing, and analysis — areas where precision and compliance are paramount.

50+ Featured Projects: A Diverse Portfolio

Out of 260+ projects, here are over 50 that showcase the breadth and depth of what's possible with AI-assisted development. These span from enterprise systems to creative experiments, hardware to cloud, each solving real problems with production-ready code.

These 50+ projects represent just 20% of my total output, yet they demonstrate the incredible range possible with AI assistance. From 42-line utilities to 188,000-line enterprise systems, each was built with the same part-time schedule that previously yielded a fraction of the output.

Looking Ahead: 2 Million Lines in 2026

Based on current trends, I expect to write well over 2 million lines of code in 2026. This isn't just linear extrapolation — AI tools are improving rapidly, and my workflows are becoming more refined. What seems ambitious today will likely seem conservative by year's end.

The implications extend beyond personal productivity. If one person working part-time can generate a million lines of functional code, what happens when entire teams adopt these tools? We're witnessing the democratization of software development at an unprecedented scale.

The Complete Project List

Below are all 260+ projects I've built since 2023, sorted by date (newest first). Each represents a unique solution, tool, or experiment. Click on any project to see its description and technical details.

All Projects (260+)

Loading projects...