Current opportunities at Zetta portfolio companies

Zetta Venture Partners
Zetta Venture Partners

ML Engineer

Guardrails AI

Guardrails AI

Software Engineering, Data Science
San Francisco Bay Area, CA, USA · San Francisco, CA, USA · United States · Remote
Posted on Tuesday, April 16, 2024

About Guardrails AI

Our mission is to unlock the opportunity offered by Generative AI by removing key barriers in the reliability and robustness of AI systems. We understand the potential of Generative AI, but we also see its limitations. Traditional software sets a high bar for reliability, and right now, LLMs aren't quite there. That's where we step in — we're passionate about extracting all the incredible possibilities Generative AI offers, and keen on empowering AI builders to get to production.

Guardrails AI manages AI volatility with guardrails that actively detect, quantify and mitigate specific types of AI risks. Multiple guardrails can be combined together to form input and output guards that protect your application from risks like hallucinations, PII and data leakage, financial advice, competitor mentions, going off topic, etc.

We're an open source company with two OSS products: Guardrails Hub (a repository of 50+ guardrails across many use cases) and Guardrails orchestration framework (framework for combining and running guards with parallelism, streaming and logging support).

We recently raised $7.5 million in a seed round led by Zetta Venture Partners with participation from Factory, Pear VC, Bloomberg Beta, GitHub Fund and angels including renowned AI expert Ian Goodfellow.

Key Responsibilities

As a senior Machine Learning Engineer at Guardrails AI, you will be contributing to a critical component of the company’s roadmap.

  • Your primary responsibilities will be to own end to end ML projects. A key component of your work will be to scope out and build whatever is the best-possible ML solution (high performance, low cost, low latency) to green-field problems. This includes scoping out the ML use case, designing and curating appropriate datasets, shortlisting algorithms, training models (including finetuning), experimentation, evaluation and deployment.

  • You’ll build general solutions to improve the reliability and robustness of LLMs, as well as industry specific solutions focusing on finance, healthcare, etc. You’ll work closely with customers on understanding their problems.

  • You’ll think about the best way to generalize your work and build abstractions to improve ML and experimentation infrastructure.

About you

  • You have minimum 4 years of experience building Machine Learning models in production settings. Your role is not just the research and experimentation in order to build a SoTA models, but also to ship the product to a production environment.

  • You’re excited about being cross-functional and owning the lifecycle of building ML solutions — from model development to deployment.

  • You’re extremely comfortable with Python, sklearn, numpy and a deep learning framework such as Pytorch / Keras / Tensorflow.

  • You stay up-to-date with the latest research in your area of interest in ML. You’re comfortable reading research papers, figuring out what the key ideas from the paper are and reproducing the results in a production environment.

  • You’ve previously worked with deep learning systems on unstructured data such as text, images, LiDAR, etc.

  • You’re really excited about Large Language Models, and excited about building applications that leverage LLMs for solving problems.

  • You enjoy fast-paced environments, and exhibit a high degree of ownership and self-sufficiency. Unstructured environments are exciting to you because they represent opportunities for growth and leadership.

ML Stack

  • Python

  • Pytorch

  • Huggingface transformers

  • Numpy

  • Sklearn

  • Pandas

  • Ray