The world of software development is rapidly evolving, and a young San Francisco-based startup is attempting to rewrite the rules. Greptile, an AI-powered tool that automates code review, is in talks to raise its Series A round at a valuation of $180 million, with Benchmark likely leading the round.
Sources say the round is expected to be around $30 million and led by Benchmark partner Eric Vishria, though terms are not yet finalised and subject to change. For a startup just emerging from stealth, that’s a significant endorsement not just of its tech, but of its timing.
A modern fix for an age-old pain point
Anyone who’s worked in software engineering knows the bottleneck that code reviews can become. They’re essential for quality assurance, collaboration, and knowledge sharing. Yet, they’re time-consuming and manual, especially in large codebases. Greptile aims to transform that dynamic.
The startup has developed a system that reads and interprets a codebase much like a seasoned engineer might. Rather than merely scanning for syntax errors or linting violations, Greptile’s AI models go deeper. It learns from the code itself, understanding design patterns, commenting on architecture, and even answering questions about logic and implementation. It is designed to be a thoughtful reviewer. Beyond code review, Greptile can also auto-generate context-aware commit messages, update documentation based on code changes, and serve as a knowledge base by integrating with documentation tools such as Quip or Google Docs.
Greptile can operate both in the cloud and on-premises, allowing flexibility for startups and larger enterprises, including those with stricter compliance or privacy requirements. While the tool can analyse any codebase via GitHub links, some reports mention that it faces challenges when processing particularly large or poorly documented repositories.
More than an AI plugin
Greptile was founded by Daksh Gupta, Soohoon Choi, and Vaishant Kameshwaran in 2023 with the support of Y Combinator, where it initially caught the eye of investors intrigued by its technical ambition. After completing YC’s Winter 2024 batch, Greptile raised a $4 million seed round led by Initialised Capital. Unlike other AI copilots or autocomplete tools focused on writing new code, Greptile is zeroing in on a different problem of how developers read, understand, and manage existing code.
Software companies today often wrestle with legacy systems, sprawling codebases, and onboarding challenges. A tool that can demystify unfamiliar code can cut hours from the typical ramp-up time for engineers and improve code quality across the board. This is where Greptile’s value proposition clicks: helping humans work faster with machines, not against them.
Greptile’s initial user base includes software teams ranging from startups to mid-sized tech firms. According to people familiar with its growth, usage has doubled in recent months, as teams deploy it not just for reviews but also as a teaching tool for junior engineers or as a knowledge base that reduces internal documentation debt.
Quiet growth, loud potential
What’s striking about Greptile is its relatively low profile until now. Unlike many AI startups that launch with bombast and bold proclamations, Greptile has moved deliberately. It prioritised refining its core engine over courting headlines, and it has largely stayed out of the marketing echo chamber that often surrounds AI tooling.
This product-focused approach may appeal to Benchmark, known for backing technically ambitious companies with deep, defensible technology. In a market increasingly overrun by ChatGPT wrappers and flashy demos, Greptile stands out by doing the hard work of building infrastructure that gets used.
However, Greptile’s work culture has drawn public attention and industry scrutiny. Founder Daksh Gupta has openly stated that the company’s work schedule runs from 9 a.m. to 11 p.m., six days a week, offering “no work-life balance,” a point that has sparked debate about startup culture and employee well-being.
Beyond code review
While its current focus is on making code review smarter and faster, insiders hint at a broader roadmap. Greptile could eventually become a central intelligence layer across the entire software development lifecycle, helping teams not just review, but plan, refactor, and even architect projects collaboratively with AI assistance.
If it succeeds, it could redefine how teams manage complexity in codebases as they scale. Instead of siloed documents and tribal knowledge, imagine a searchable, conversational memory of how and why every line of code came to be.
Final thoughts
In an age where developers are under pressure to ship faster and maintain higher standards, tools like Greptile feel inevitable. The startup’s early traction, coupled with Benchmark’s likely backing, suggests that investors see it as more than a neat AI experiment.
For now, Greptile remains in limited beta, quietly improving behind the scenes. But if it plays its cards right, it may soon become the kind of behind-the-scenes infrastructure that powers the software we all use without most of us ever knowing it.
