LLMs have the tendency to hallucinate. And that's because they're only as knowledgeable as the data they're trained on. Our approach to generating documentation is deterministic, prioritizes reliability, and involves 3 steps: 👆 Static analysis identifies relevant flows & logical components ✌️ Deterministic retrieval of relevant context for specific topics or questions 🤟 We only introduce LLMs at the end, transforming the accurate context into documentation. Read more:
Swimm
Software Development
New York, NY 8,318 followers
Understand & work with any codebase in minutes | 2024 Gartner Cool Vendor
About us
Swimm is an AI-augmented platform that helps enterprise developers quickly understand big, complex codebases—and seamlessly captures knowledge to fill in any documentation gaps.
- Website
-
https://www.swimm.io
External link for Swimm
- Industry
- Software Development
- Company size
- 11-50 employees
- Headquarters
- New York, NY
- Type
- Privately Held
- Founded
- 2019
- Specialties
- Software Documentation, Code Documentation, DevTools, Developer Tools, Developer Productivity, software engineering, AI, ai coding tools, and ai developer tools
Locations
-
Primary
200 Park Ave S
New York, NY 10003, US
-
Daniel Frisch 3
Tel Aviv, Israel, IL
Employees at Swimm
Updates
-
DORA began researching code documentation‘s impact on the software development lifecycle in 2021. And ever since, their findings have consistently shown that high-quality documentation significantly impacts organizational performance. Here's a breakdown of the key documentation takeaways from the 2024 report→
Here’s what the 2024 DORA report has to say about code documentation - Swimm
https://swimm.io
-
Say hello to the Swimm for GitHub Copilot extension! 🚀 Turn Copilot Chat outputs into detailed documentation that is always up to date. Plus, each new document generated enhances Copilot's context, aligning outputs with your organization's standards and logic. Read about it here → https://lnkd.in/dXA8TiAV
-
We just wrapped up two amazing days at GitHub Universe! It was so exciting to be back for our second year, meeting devs from all over the world, and deepening our partnership with the team at GitHub. We were especially excited to launch something new: the Swimm GitHub Copilot Extension. With our new extension, you can create documentation directly from Copilot Chat. Want to see it in action? Gilad Navot will be joining a GitHub webinar to officially launch the extension. Sign up here → https://lnkd.in/djWdesWq
-
Working with legacy code doesn't have to be such a daunting task. In fact, understanding what you have is the first step in making legacy systems more accessible for the whole team. We teamed up with LeadDev for a webinar where engineering leaders will give you proven strategies for tackling legacy systems. Join us on Oct 30 at 13:00 EDT/17:00 GMT. Save your seat→ https://lnkd.in/du678yEq
A roadmap to working with your legacy codebases
leaddev.com
-
Thanks for having us on the show, Simon Maple and Tessl! On this episode of The AI Native Dev, Omer Rosenbaum and Simon broke down the best ways to leverage AI for code documentation, especially when it comes to legacy languages like COBOL. Listen/watch the entire episode here (or wherever you get your podcasts!)→ https://lnkd.in/earhGr6f
-
"Swimm and Backstage from Spotify gave us the documentation backbone we needed to support our scaling engineering organization" - Ryan McKenna, Staff Engineer at RVO Health By integrating Swimm and Backstage, the team at RVO Health was able to ship faster, improve team collaboration, and benefit from better quality documentation. 👀 Read the full case study here → https://lnkd.in/dSUwPxpg
-
Swimm reposted this
Had a really insightful discussion around where AI meets documentation with Omer Rosenbaum, CTO and Co-founder at Swimm. We covered a wide range of topics around the need for documentation, why drift occurs between code and docs, why devs avoid writing docs and much more! Here are some of the questions we discussed: * Is it possible for AI to fully automate documentation creation, or will human input always be necessary? * How does the balance between static analysis and AI impact the quality and accuracy of code documentation? * How can organizations ensure that documentation, especially for legacy codebases, remains a reliable source of truth amidst constant code changes? #AIDocumentation #AINative #AINativeDev #AINativeSoftware #Swimm #OmerRosenbaum And one of my favourite quotes of the session:
-
Repeat after us: Working with legacy code doesn’t have to suck. Our latest white paper will help ensure that legacy code complexity never blocks your projects again. Learn how Swimm automatically generates documentation for entire repositories– regardless of programming language. Download it here → https://lnkd.in/dhutTgzy
Auto-documenting legacy code with Swimm
https://swimm.io