Tusk offers a way to ship fast while improving code quality, with its AI agent generating unit and integration tests, using the user’s codebase context and business logic
By Nileena Sunil
Y Combinator-backed startup Tusk is an AI agent that generates unit/integration tests for pull requests (PRs). Founders Marcel Tan and Sohail Kshirsagar are the team behind the first publicly available coding agent, and they have now turned their attention to another software engineering task—writing automated tests.
Tan and Kshirsagar believe that while writing tests is important for maintaining high quality for code, most software engineers who need to hastily meet deadlines see writing the unit and integration tests as the “first thing to go.” This often causes problems later, with customers complaining about bugs in the feature.
READ: Formula Insight unveils new platform to simplify equity research (December 6, 2024)
Founded in 2023, Tusk offers a way to ship fast while improving code quality, with its AI agent generating unit and integration tests, using the user’s codebase context and business logic. Tusk functions as a non-blocking PR check that suggests happy path and edge case tests that have not been covered by the existing test suite. The AI agent shows users which tests have passed or failed and allows them to incorporate the test cases into the PR with one click.
Headquartered in San Francisco, Tusk has been used by engineering companies like DeepLearning.AI, Wyndly, and Afterword, to prevent bugs while increasing test coverage.
Afterword’s Co-Founder and CTO Zack Moy said, “Tusk is the first AI tool that noticeably accelerated our dev cycle. By scaling our testing infra, it’s given us confidence to ship faster and been instrumental in helping our lean team scale across 3 stacks. Of all the AI tools being built right now, I’m most excited about Tusk.”
While Senior Backend Engineer at DeepLearning.AI Amo Chen said, “Edge cases that are not covered can cause the most critical issues. By using Tusk, our team gains an additional guardian to protect our product from the threats of edge cases.”
Aakash Shah, founder and CEO at Wyndly noted, “Our priority as a fast-shipping product team is making sure we get high quality code out the door quickly. Tusk has been instrumental in helping our engineering team increase test coverage with minimal effort.”
Tusk is also developer-centric, and fully customizable. It puts engineers in allowing them to accept or decline the generated tests, and customize configs to meet teams’ testing guidelines.
According to Tan and Kshirsagar, Tusk surpasses other test generation and code review tools by running the tests it generates in an ephemeral sandbox and reiterates on its output to ensure tests are checking for happy paths and edge cases. It is also optimized for test generation quality, with 69% of Tusk-generated suits incorporated into customer’s PRs.
READ: Y Combinator startup launch: Unlock AI-driven user assistance with Layup (December 10, 2024)
Tan, the co-founder and CEO of Tusk, is a self-taught software engineer with a proven track record in driving growth. As a former technical product manager for 6sense’s AI email product, he scaled the product’s total contract value from $300,000 to $8.9 million in just one year. A graduate of the University of California, Berkeley, he has been collaborating with co-founder Kshirsagar on projects since their college days.
Kshirsagar, the co-founder and CTO of Tusk, is focused on empowering engineers to spend more time on meaningful work. Before Tusk, he was a senior software engineer at Aspire, where he led the workflow orchestration and automations team, driving millions of influencer collaborations.