menu
Refrakt

Refrakt

●1 ●5 ●11
calendar_today • schedule1 min read

Deep learning workflows today are fragmented across scripts, dashboards, and tools.

Refrakt is a unified platform that brings model training, evaluation, and explainability
into a single system.

This launch includes a short 3-minute UI walkthrough (no live execution yet): http://youtube.com/watch?v=IZQ8kW2_ieI

Platform access is coming soon!

Any feedback is welcome!

2 Comments

•

Interesting idea bringing training and explainability into one place, curious how Refrakt will handle collaboration when multiple people are iterating on the same model.

1 vote
•

@[Andrew Mewborn] great question!

Refrakt doesn’t treat “the model” as a shared mutable object. The primary units are
versioned workflows, configs, and the artifacts produced by each run.

Collaboration is centered around reproducible runs and explicit artifacts (e.g.
being able to inspect, compare, or reuse outputs from prior executions) rather than
multiple people concurrently editing the same state.

For now, the focus is on making iteration legible and traceable (who ran what, with
which config, producing which artifacts).

More explicit team-level collaboration primitives are something we’re thinking about, but this walkthrough is mainly meant to show the core workflow abstraction.

1
🔥 Join developers growing publicly
Share your knowledge, build in public, and grow your developer presence with a global community.
Join CoderLegion

More Posts

Everyone says DeepSeek is cheaper, but I got tired of guessing the exact math. So I built a calculat

abarth23 - Apr 27

I’m a Senior Dev and I’ve Forgotten How to Think Without a Prompt

Karol Modelski - Mar 19

The Sovereign Vault — A Comprehensive Guide to Protocol-Driven AI

Ken W. Algerverified - Jun 4

Your AI Doesn't Just Write Tests. It Runs Them Too.

Kevin Martinez - May 12

You Don't Need a Ministry of Truth to Build a Memory Hole

Ken W. Algerverified - Aug 19
chevron_left
Akshath Mangudi
181 Points • 17 Badges
akshath.tech
3Posts
3Comments
2Connections
ML systems engineering.

More From akshath

Weightlens

Refrakt Demo

Related Jobs

View all jobs →

Commenters (This Week)

2 comments
2 comments
1 comment

Contribute meaningful comments to climb the leaderboard and earn badges!