Skip to content

Commit

Permalink
updated interfaces overview page.
Browse files Browse the repository at this point in the history
  • Loading branch information
djl11 committed Feb 19, 2025
1 parent 72689d9 commit 54fba82
Showing 1 changed file with 17 additions and 16 deletions.
33 changes: 17 additions & 16 deletions interfaces/overview.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -2,26 +2,27 @@
title: 'Overview'
---

Interfaces are **fully hackable**, which you can use to build *personalized* pipelines
for: logging, evaluations, guardrails, human labelling, agentic workflows,
self-optimization, and more.
Interfaces are **fully hackable**, which you can use to build *custom* interfaces for:
logging, evals, guardrails, labelling, tracing, agents, human-in-the-loop, hyperparam sweeps,
and anything else you can think of ✨

simply `unify.log` your data, and then compose your own custom interface using the
three core building blocks: (1) **tables**, (2) **plots** and (3) **views**.
Just `unify.log` your data, and add an interface using the four building blocks:

Despite the explosion of LLM tools, many of these are inflexible, overly abstracted, and
complex to navigate.
1. **tables** 🔢
2. **views** 🔍
3. **plots** 📊
4. **editor** 🕹️ (coming soon)

Tooling requirements constantly change across *projects*, across *teams*, and across
*time*. We've therefore made Unify as simple, modular and hackable as possible, so you
can spin up and iterate on the *exact* AI platform that **you** need, in **seconds**
Every LLM product has **unique** and **changing** requirements, as do the **users**.
Your infra should reflect this!

Rather than going through *each concept* in detail, we explain interfaces through a
simple *case study*, where we're building an education app for students.
<div style="display: block;" align="center">
<a href="https://console.unify.ai/">
<img class="dark-light" width="100%" src="https://raw.githubusercontent.com/unifyai/unifyai.github.io/refs/heads/main/img/externally_linked/notion_for_llm_observability.gif"/>
</a>
</div>

This single case study walks through the use of **logging**, **evaluating** and
**debugging** the LLM application. This is by no means an exhaustive list (you can
build an interface for pretty much *anything*), but it should give you an idea of
how flexible and hackable interfaces are 🛠️
We've tried to make Unify as **(a) simple**, **(b) modular** and **(c) hackable** as possible,
so you can quickly probe, analyze, and iterate on the data that's important for **you**, your **product** and your **users**

Head over to the next page to dive in! ➡️

0 comments on commit 54fba82

Please sign in to comment.