About this software name generator
A software product name commits to a category and an ecosystem before it commits to anything else. 'Glint' is a B2B SaaS QA tool. 'kestrel' is an open-source Rust HTTP client. 'tw' is a two-letter CLI utility. Each name is shorter than its concept and longer in implication than its letters. Most online software-name generators produce generic decorative phrases ('Synergyzz,' 'Codinatorr') with no category-fit and no ecosystem awareness. This software name generator doesn't, and that is what it is built for.
Each result draws on real software-naming tradition — the short-and-confident SaaS tradition of Linear, Notion, and Figma; the Rust / Python / JavaScript open-source library tradition; the CLI utility tradition of ripgrep, fzf, jq, and gh; the data-infrastructure tradition of Snowflake, Clickhouse, and DuckDB; the AI / ML tradition of Anthropic, Hugging Face, and Replicate.
What you get
Each result returns a software name, an etymology and category-fit description, the principal language or runtime (if applicable), the closest existing comparable in the same category, a plausible maintainer-or-founder backstory (how the name was chosen, when first released, license), a documentation-and-brand-voice paragraph, and a usable one-liner suitable for the README or the homepage.
The categories the generator rotates
B2B SaaS app — Linear, Notion, Figma, Loom, Airtable.
Developer-facing platform / IaaS — Vercel, Render, Fly.io, Cloudflare, Heroku.
Open-source library — React, Vue, Svelte, Django, Flask, Rails, Prisma.
CLI utility — ripgrep, fzf, bat, fd, jq, gh, kubectl.
Database / data infrastructure — Snowflake, Postgres, Clickhouse, DuckDB.
AI / ML platform — Anthropic, OpenAI, Hugging Face, Replicate, Modal.
Security / infrastructure — Snyk, Tenable, CrowdStrike, Datadog, Sentry.
Productivity / consumer app — Things, Bear, Drafts, Day One, Reflect.
Game-dev tool / creative software — Godot, Aseprite, Blender, Krita.
Internal-platform / framework — Bazel, Buck, Pants, Lerna, Turbo, Nx.
What a good software name has to survive
Software names face tests other brand names do not. The first is the namespace: the perfect name is worthless if the npm package, the PyPI project, the crates.io crate, or the Homebrew formula is already taken, so availability on the relevant registry is a hard gate rather than a nicety. The second is searchability. A name that is also a common English word can be almost impossible to find in a search, which is the quiet reason so many tools reach for a slightly-invented spelling (Vercel, Heroku, Svelte) over a plain dictionary word. The third is the type-test: a CLI users invoke fifty times a day wants to be two or three letters, which is why CLI names trend terse and lowercase while a SaaS product, typed once at signup, can afford a fuller word. The generator weighs all three, so the name it hands you is one that could actually ship.
How to use these names
For real software-naming, generate 10–15 candidates, then run each through the standard four-filter check: (1) the package-manager identifier is available on the relevant registry (npm / pip / cargo / brew / Docker), (2) no major existing project in the same ecosystem already uses the name, (3) the name is pronounceable on first hearing in English, (4) the brand voice and one-liner are consistent with the category position.
For fiction — a startup in a novel, a fictional library a hacker character uses, a megacorp's internal platform in a cyberpunk RPG — the names and the implied tooling plug in directly with category-correct flavor.
Why category-fit is the whole game
A 'clever' software name that fights its own category is a name that buyers, contributors, and downstream users will quietly avoid. The generator commits each name to a category at generation time: SaaS candidates sound like SaaS, OSS-library candidates sound like libraries, CLI candidates sound like CLIs. That single decision — made first, made well — is what separates a working software name from a forgettable one.