Integrating Batch Changes with GitLab Issues and Epics, provides a complementary set of features to manage massive code-based changes for large organizations.

Command menu UX improvements, JSON export, JSON format changes to custom commands, autocomplete rate limit visibility, and streamed autocomplete responses.

Sourcegraph experienced a security incident on August 30, 2023. This blog describes the details, timeline, and impact of the incident.

Faster autocomplete, improved recipes (now commands), new chat layout, improved code smell detection, custom commands, and more.

We are super excited to share the new Cody features we will be releasing on June 28th, 2023 that we decided to have a giveaway.

Announcing Sourcegraph 5.1. The latest release includes updates to the code graph, tools and introduces Cody, our AI coding assistant, to enterprise instances.

Stevey’s “Cheating With Cody” blog series, Episode 3

The June 2023 release of Cody makes it more powerful and accurate with better autocomplete, new and improved recipes, multi-repository context, inline chat, support for JetBrains IDEs, and a new Cody desktop app. Cody is free forever for devs on both public and private code.

You can now assign file and repository ownership directly from the Code Search interface. Sourcegraph also provides signals, such as recent contributors, to help you infer file ownership and save time.

GitLab customers can leverage Cody to tap into the full knowledge of their codebase, no matter the size, with a simple chat interface.

Sourcegraph now supports cross-repository precise code navigation for C and C++ for code indexed by scip-clang.

We’ve increased the completions rate limit to 1,000 completions per day for Sourcegraph.com users.

scip-clang is our new SCIP indexer for C and C++ supporting precise code navigation in Sourcegraph, supporting a wide variety of language features.

Stevey’s “Cheating With Cody” blog series, Episode 2

Cody is now available to anyone with a Sourcegraph.com account.

AI is the best and worst thing to happen to enterprise dev teams.

Embrace efficient programming tools and practices, overcome reluctance to change, and utilize newfound time for productive tasks in software development.

The advent of LLMs has the potential to revolutionize the way we produce and consume information, but it is important to be realistic about the hype and focus on delivering real value.

We're eager to see Twitter's algorithm go open source. Sourcegraph devs (and our Discord community) will be liveblogging the most interesting things we find.

If this LLM is so damn smart, why can't it count how many fingers I'm holding up?