Open source AI copilot that knows your company's code and its context
With AI increasingly automating away code generation, software engineers will spend more time reading, judging, and architecting code rather than writing it. Storia is building an open-source copilot that knows a company's codebase and its context. We are starting with Sage, a Perplexity-like agent for helping developers understand, judge, and generate software. Given an existing codebase, developers can ask Sage questions such as: 1) Given my project’s SLA and latency constraints, what is the appropriate underlying vector database to use? How would I incorporate it into my existing codebase? 2) Why should I pick Redis over Milvus as my underlying vector store? 3) Does this codebase in our organization still work and what steps are required for a complex integration with another library? Sage’s answers are directly supported by documentation and external references like GitHub, Stack Overflow, technical design documents, and project management software, preventing hallucinations. Today, Sage has up-to-date knowledge about open-source repositories (indexed daily). Tomorrow it will have a deep understanding of every line of code on the Internet. For teams, Sage will know about your private codebase too. No group has yet solved how to build an AI system that comprehends a codebase and its context and can empower every developer to architect better code, faster. This requires new research advances because vanilla RAG and out-of-the-box LLMs aren’t going to cut it. We have 20+ years of software engineering and AI research experience. Julia worked on precursors of Gemini using contextual neural techniques before they were called “RAG” (and applied it to products like Google Keyboard and Pixel phones). Mihail built the earliest LLMs at Amazon Alexa and launched the first contextual deep learning conversational AI system in production at Alexa.
Active Founders
Julia Turc, Founder
Cofounder @ Storia & previously at Google Research. 10+ years in NLP. Built the first bidirectional RNN for machine translation at Oxford. At Google, I helped build pivotal technologies like federated learning, the AI layer on top of Android, and RAG for precursors of Gemini. Excited about how NLP is now redefining software engineering.
Company Launches
Hi everyone. We’re Julia and Mihail, the team behind Storia. We have 20+ years of software engineering and AI research experience. Julia worked on precursors of Gemini using contextual neural techniques before they were called “RAG” (and applied it to products like Google Keyboard and Pixel phones). Mihail built the earliest LLMs at Amazon Alexa and launched the first contextual deep-learning conversational AI system in production at Alexa.
👎 The Problem: <5% of software teams use AI code understanding and generation systems
Despite all the hype around AI coding following the release of GitHub Copilot, a disappointingly small percentage of engineering teams have actually adopted AI assistants into their developer workflows. Why?
What developers don’t like about systems today:
- Code search is not accurate enough, slowing dev velocity.
- Code assistants are not grounded in their team’s codebase, so they lack the necessary context to produce useful and high-quality answers.
- Code outputs can be incorrect due to hallucination which strains dev patience and trust.
🚀 The Solution: A contextual AI pair programmer built for any team’s codebase
We’re building Sage, a Perplexity-like agent for helping developers understand, analyze, and generate software. Given an existing codebase, developers can ask Sage questions such as:
- For my project’s SLA and latency constraints, what is the appropriate underlying vector database to use? How would I incorporate it into my existing codebase?
- Why should I pick Redis over Milvus as my underlying vector store?
- Does this codebase in our organization still work, and what steps are required for an integration with another library?
A few of the features Sage supports:
- We index repos daily, providing users with the most up-to-date info about how to use a library (no stale, deprecated API definitions),
- We have a Perplexity-like focus on citing the source of every generated line of code, enhancing developer trust.
- We've developed algorithms grounded in research for syntactic and semantic understanding of a codebase so that we can provide more relevant context and use it effectively given a user query.
We are now actively working with seed to series C partner companies to help them integrate Sage into their developer workflows.
🌅 The Opportunity: Making every engineer on a software team a 10x developer
There are ~28M developers worldwide. Github Copilot is the frontrunner in AI assistant tooling and reached $100M ARR in 2 years, which is only a small percentage of the $913B spent on software development in 2023.
🙏 Our Ask
We’re still onboarding a closed group of beta partners to integrate with Sage. We would appreciate connecting with engineering leaders (VP of engineering, Head of Engineering, CTO) at Seed-Series C companies.
Additionally, if you want your open-source repository indexed reach out to us at mihail@storia.ai.
YC Sign Photo
Company Photo
