What Sourcegraph shipped
Sourcegraph released a new integration connecting their Deep Search capability directly to Claude Code. This feature allows platform engineering teams to query massive, distributed codebases without leaving the terminal.
By bringing enterprise-grade search into an AI coding assistant, developers can locate deprecated dependencies and review search paths in a unified workflow. This reduces context switching and maintains secure codebase visibility for engineering leadership.
How the motion works
When demonstrating terminal-based workflows, motion design must balance legibility with momentum. A standard screen recording often feels static and visually sparse. To elevate this, designers can use a punch-in to direct focus toward specific command line inputs, cropping out the visual noise of an empty terminal window.
Applying a typewriter effect to command inputs or AI-generated responses helps simulate the live, interactive nature of the tool. Rather than dropping blocks of text onto the screen instantly, pacing the text reveal matches the cognitive rhythm of a developer reading the output in real time.
Managing hold-time is critical when showing complex code queries. The viewer needs enough time to parse the search parameters and the resulting output. Pausing the action just after a search returns results ensures the audience grasps the scale of the codebase being queried before the video transitions to the next step.
What to steal from it
- Use punch-ins to isolate terminal commands and eliminate unnecessary screen real estate.
- Pace text reveals to match the natural reading speed of a developer processing code.
- Ensure hold-times on complex search results are long enough for viewers to register the data.
- Maintain a consistent grid and typography scale to keep terminal demonstrations grounded.
Want a video like this for your product?
Impractical cuts one from a single prompt — the same motion craft, in about twenty minutes.
