Home
Companies
Atla

Atla

We build LLMs to evaluate other LLMs

We're building the world's most capable AI evaluation models & the tools to unlock the full potential of language models for developers.

Jobs at Atla

London, England, GB
£100K - £150K GBP
6+ years
London, England, GB
£100K - £160K GBP
3+ years
London, England, GB
£100K - £160K GBP
3+ years
London, England, GB
£100K - £160K GBP
3+ years
Atla
Founded:2023
Team Size:5
Location:London, United Kingdom
Group Partner:Harj Taggar

Active Founders

Maurice Burger

Co-founder & CEO of atla (S23). Startup veteran @ Syrup, Trim, and Merantix. Masters in CS @ University of Pennsylvania. Half an MBA @ Harvard Business School.

Maurice Burger
Maurice Burger
Atla

Roman Engeler

Co-founder & CTO of atla (S23). AI safety researcher @ MATS. MSc. Robotics @ ETH, Stanford, Imperial.

Roman Engeler
Roman Engeler
Atla

Company Launches

At Atla, we train models to act as reliable evaluators of generative AI applications. Our evaluation models are faster, more stable, and align more closely with human annotators than general-purpose LLMs.

The Problem

Benchmarks ≠ user preferences. Developers currently spend huge amounts of time iterating across prompts, models & training data. The only ways to evaluate performance are manual & subjective or slow & unstable. You either trust your gut, hire expensive domain experts, or have to live with the self-bias and instability of GPT-4-based evals.

Our Solution

With Atla, teams can rapidly achieve high performance, discover failure modes, and know the accuracy of their GenAI applications. Our evaluation models are faster, more stable, and align more closely with human annotators than general-purpose LLMs. Atla helps engineers get a clear optimization target for their LLM applications. 🎯

Our 7B eval model compares favorably with GPT-4 on our experiments with real customer data and beats GPT-3.5 as an evaluator on the most important benchmarks:

We have trained on five key metrics to cover the most critical use cases of LLMs:

hallucination: Measures the presence of incorrect or unrelated information.

groundedness: Measures if the answer is factually based on the provided context.

context_relevance: Assesses the retrieved context’s relevance to the query.

recall: Measures how completely the response captures the key facts.

precision: Assesses the relevance of all the information provided.

How it works

  1. Integrate Atla into your codebase. Use our API as a drop-in replacement for OpenAI.
  2. Receive reliable scores & critiques that are reproducible.

Using OpenAI's SDK, the following

becomes

Here’s a sample workbook with human-annotated data from when we built a legal co-pilot. Feel free to try it on your own data!

Our Ask

Have an LLM in production working on generative tasks? Try out our model to evaluate how well your AI application is really working:

  1. Fill in this quick onboarding form (< 30s) to receive a private API key. We offer $20 of free credits to get you started.
  2. If you want a demo or want to chat more about your evals, please book a call here.

Other Company Launches

atla - honest AI assistants for legal teams

atla helps in-house lawyers get reliable answers without billable hours
Read Launch ›