Learn how Draup documented 10,000 function lines of code in just one week
Problem
Documenting and refactoring Draup’s 1.5 million-line codebase manually would take months and stall development.
Solution
CodeAnt AI auto-generated 10 k docstrings and fixed 1.2 k issues in a week, cleaning the entire codebase at scale.
Product
Founded
2017
Funded
$20 Million
Code Hosting
GitHub
Tech Stack
Python, Java, JavaScript, and Node.js Docker
Industry
SaaS
About
Draup is the go-to AI-driven Sales Intelligence solution for Fortune 500 enterprises, boasting a client list that includes Microsoft, Shopify, Pepsico, Nokia, T-Mobile, and more. Every day, they analyze a whopping 1 Billion data points from over 50,000 sources, providing invaluable insights for HR and Sales Leaders.
Challenge
Draup recognized the need to enhance their coding practices and decided to kickstart this by cleaning up their entire codebase. The CTO of Draup, Kashish was on the lookout for a tool that could efficiently clean their entire Python codebase, especially with an upcoming week dedicated to company-wide code cleaning in mid-December.
They had specific requirements:
Generate standardized and context-driven code documentation
Detect and auto-fix anti-patterns related to resource usage, latency, and readability
Detect and suggest refactoring for dead and duplicate code
Find complex functions and suggest auto-refactoring
Draup sought a comprehensive tool to enforce clean coding practices throughout the entire development process, starting from the shift-left principle and extending to developers' IDEs, Continuous Integration, and Pull Request checkers.
Ensuring adherence to best practices at these three crucial touchpoints minimizes time spent on triaging and fixing bugs.
Solutions
Upon connecting with CodeAnt AI, Kashish, the CTO of Draup, discovered that our offerings perfectly aligned with their requirements. Leveraging all three of our integrations:
VS Code & PyCharm Extensions: Facilitated clean coding practices directly in developers' IDEs
CodeAnt AI Dashboard: Empowered bulk fixing of up to 200 files efficiently
PR Reviewer : Enabled auto-fixing on every change in Pull Requests
CodeAnt AI analyzed 1.5 million lines of code, documented 10,000 functions, and auto-fixed 1,200 issues, doing three months of work in just one week—while continuously reviewing hundreds of new code commits daily.
Alpesh Jain
VP Engineering at Good Glamm Group
Result
CodeAnt AI deployed its AI model and rule-based engines in Draup's AWS infrastructure, thus ensuring that no data will ever leave their infrastructure, and here are the results for the same.
Scanned over 1.5 Million lines of code
Generated more than 10,000 standardized, context-driven code docstrings
Detected and Auto-Fixed 1,200 antipatterns related to resource usage, latency, and readability issues
Fixed 3,000+ complex functions, streamlining the codebase for better maintainability
Detected security vulnerabilities, providing an ability to address potential threats promptly
CodeAnt AI's integrations not only enhanced Draup's coding practices but also significantly contributed to the overall quality, security, and efficiency of their software development process.
