Machine Learning Engineer

Onyx Security; San Francisco (CA); On-site; USD 130,000 - 160,000; Full time

Job summary

Onyx Security is hiring an AI/ML Engineer to enhance their open-source project. You will be working in San Francisco, focusing on improving LLM capabilities and knowledge retrieval. The role involves collaboration with founders to shape product direction.

Ideal candidates should have 3+ years of experience in AI/ML, strong software engineering skills, and familiarity with relevant technologies like PyTorch and NLP models. There are also bonus points for full stack skills and interest in technical writing.

Qualifications

• 3+ years of AI/ML engineering experience building real-world applications.
• In-depth experience with PyTorch/Tensorflow, NLP models, and standard ML algorithms.
• Strong software engineering background capable of building backend features.

Responsibilities

• Evaluate and implement LLM based knowledge graphs and algorithms.
• Improve platform user experience through feedback and personalization.
• Own the approach from inception to production code.

Skills

AI/ML engineering experience

PyTorch/Tensorflow

NLP models

Web frameworks

Communication skills

Tools

Typescript

React

NextJS

Python

Postgres

Job description

About the role

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 The project has over 10K stars and over 3K community members across Slack and Discord (these stats may already be out of date when you read this). You’ll have the opportunity to build in the open and your work may be used by millions of people in the future.

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.

Your impact

You’ll 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 users’ 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

You’ll be successful if you…

• 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 to 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
• Excellent communication skills and ability to collaborate with full stack roles

Bonus points

• Familiar with the full stack Typescript/React/NextJS, Python, Postgres
• Interested in writing technical blogs to establish Onyx as a leader in the space

About the interview

• Non-technical Phone Screen (30 mins)
• ML Interview (45 mins)
• Practical Coding Interview (30 mins)
• Work Trial (3 days in person, fully covered + compensated)