/AI

Building Personal Software With Claude

- Nelson Elhage tl;dr: “This experience has shifted a bunch of my thinking about the role of LLMs in software engineering and in my own work. These thoughts are still unfolding, but this piece is an attempt to capture my experience, and to think aloud as I ponder how to update my behaviors and beliefs and expectations.”

featured in #587

AI Coding Agents For Engineering (And Business) Impact

tl;dr: There’s a lot of BS about AI coding agents. Sourcegraph’s AI coding agents actually work. Our code review agent uses specific rules you define, instead of trying to replace humans entirely. They use search + AI to help you define rules precisely and eval against recent PRs.

featured in #586

How Might AI Change Programming?

- Thorsten Ball tl;dr: Thorsten poses questions about future implications of AI: Will this affect programming language adoption? Will code optimization shift to focus on AI readability? Could prompts replace stored code? Will we need new ways to handle AI-generated technical debt? 

featured in #586

Judging Code

- Thorsten Ball tl;dr: Thorsten creates a simple system where an LLM evaluates website code based on must-have and nice-to-have requirements, scoring it from 0-5. He demonstrates its reliability and consistency, suggesting that LLMs could replace traditional code-based approaches for certain evaluations.

featured in #585

GenAI: Perception vs Reality

tl;dr: In 2024, Jellyfish introduced the Copilot Dashboard to measure the impact of the most widely adopted genAI coding tool. We’ve since gathered data from over 4,200 developers at more than 200 companies, giving us a representative sample of how engineering organizations are using Copilot and what impact it’s having on production.

featured in #579

How To Cut Developer Ramp Time With AI

tl;dr: Developers must ramp up and navigate the new complex codebase of a new customer every quarter. Architects need to do that every month. Learn how they transformed this challenge into a competitive advantage using deep context AI.

featured in #578

How To Cut Developer Ramp Time With AI

tl;dr: Developers must ramp up and navigate the new complex codebase of a new customer every quarter. Architects need to do that every month. Learn how they transformed this challenge into a competitive advantage using deep context AI.

featured in #577

AI-Generated Tools Can Make Programming More Fun

- Geoffrey Litt tl;dr: “I want to tell you about a neat experience I had with AI-assisted programming this week. What’s unusual here is: the AI didn’t write a single line of my code. Instead, I used AI to build a custom debugger UI… which made it more fun for me to do the coding myself.”

featured in #577

Autonomous Coding: Are We There Yet?

tl;dr: BlueOptima's study of 110k developers and 82 million code changes provides a coding automation framework inspired by the SAE Driving Automation levels. It shows that whilst we are currently very far from full automation the rapid rise of LLMs since 2022 requires preparation for new coding standards, hybrid expertise to mitigate quality and security issues and tailoring of automation to specific needs to maximise impact.

featured in #574

Introducing The Prompt Engineering Toolkit

tl;dr: “To facilitate rapid iteration and experimentation of LLMs at Uber, there was a need for centralization to seamlessly construct prompt templates, manage them, and execute them against various underlying LLMs to take advantage of LLM support tasks. To meet these needs, we built a prompt engineering toolkit that offers standard strategies that encourage prompt engineers to develop well-crafted prompt templates.”

featured in #572