Finance ops · Production

Tide achieves 2x faster ML model deployment with Tecton feature platform, unlocking 7x more features and 2x more models

The problem

Tide's in-house feature store created bottlenecks that made it difficult to add features, caused unpredictable production changes, and produced slow iteration times with heavy engineering requirements, resulting in model deployments that took two to four months.

First attempt

Tide's in-house feature store was unable to support the pace of real-time ML model development, producing unpredictable production changes and requiring heavy engineering resources.

Workflow diagram · grounded in source
1
Feature definition
trigger
“users can easily define and test features from diverse data sources (including streaming, batch, and real-time data)”
2
Pipeline construction
integration
“Tecton efficiently constructs and executes reliable data pipelines and infrastructure, ensuring the generation and storage of computed features for precise models and predictions”
3
Centralized feature store
integration
“Tecton's enterprise feature store is a centralized repository for managing, sharing, and retrieving features across various data sources and projects”
4
Real-time feature serving
output
“Tecton's unified serving system is designed to serve any model for any performance SLA, with a single API endpoint that provides fast, reliable, and scalable access to features needed for inference”
5
Feature reuse and backtesting
feedback_loop
“Tide benefited from the easy reuse of high-quality features, streamlined backtesting processes, and reduced deployment times”
Reported outcome

With Tecton, Tide cut model deployment time in half, deployed 2x more models, and achieved 7x more features with increased model accuracy, significantly accelerating time-to-value for new features.

Reported metrics
Model deployment timecut model deployment time in half
Time-to-production speed2x faster
Number of features deployed7x more features
Models deployed2x more models
Show all 6 reported metrics
model deployment timecut model deployment time in half
time-to-production speed2x faster
number of features deployed7x more features
models deployed2x more models
previous model deployment durationtwo to four months
ML team time on feature and data engineeringupwards of 70%
Reported stack
TectonPython
Source
https://mlops.community/blog/how-tecton-helps-ml-teams-build-smarter-models-faster
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

With Tecton, Tide cut model deployment time in half, deployed 2x more models, and achieved 7x more features with increased model accuracy, significantly accelerating time-to-value for new features.

What tools did this team use?

Tecton, Python.

What results were reported?

Model deployment time: cut model deployment time in half; Time-to-production speed: 2x faster; Number of features deployed: 7x more features; Models deployed: 2x more models (source-reported, not independently verified).

What failed first in this deployment?

Tide's in-house feature store was unable to support the pace of real-time ML model development, producing unpredictable production changes and requiring heavy engineering resources.

How is this finance ops AI workflow structured?

Feature definition → Pipeline construction → Centralized feature store → Real-time feature serving → Feature reuse and backtesting.

Related finance ops cases