About the role
đ„Â Your impact
Onyx is a popular open source project with hundreds of thousands of users. The project has over 10K stars and over 3K community members across Slack and Discord (these stats may already be out of date when you read this). Youâll have the opportunity to build in the open and your work may be used by millions of people in the future.
đĄÂ About the role
Onyx is the knowledge layer on top of LLMs. Help us improve our agent and knowledge retrieval capabilities to push the frontier on unsolved problems like multi-hop QA, needle in haystack, aggregation type RAG, etc. This is an in-person role based in San Francisco, CA.
Youâll be:
- Evaluating and implementing LLM based knowledge graphs, advanced RAG approaches (StructRAG, etc.), LLM agents, advances in NLP, multi-modal transformers, advanced information retrieval algorithms
- Working on usersâ experience with the platform through features like learn from feedback, search personalization, SME suggestion, etc.
- Build a semantic and programmatically useful understanding of the organization's priorities, projects, and people as additional signals to the answering capabilities of Onyx
- Own the approach from inception to validation to production code
- Collaborate with Founders and the Head of AI to shape and influence the direction of the product and contribute to the AI/ML engineering strategy
đ Youâll be successful if youâŠ
- Have 3+ years of AI/ML engineering experience building real-world applications
- have in-depth experience with PyTorch/Tensorflow, NLP models, and standard ML algorithms
- Are up date with new advances such as open source/proprietary LLMs, RAG and agent-frameworks
- Strong software engineering background and capable of building backend features with web frameworks, ORMs and relational DBs
