endorsedCompleted

Machine Learning Optimization for CellarEye Wine-Cellar System

Services Covered:

Custom Software Development • AI and Machine Learning

Industry

Information Technology

Duration

1 month

Budget

Confidential

Start Date

1 April 2020

Team Size

1-5 Employees

Engagement Model

Offshore

Project Capability Score (PCS)

79/100

Good

Project Capability Score (PCS) estimates how capable Xmartlabs is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

CellarEye

San Francisco, United States of America

Project Summary

Xmartlabs optimized machine learning models for a wine-cellar inventory system, experimenting with different approaches and PyTorch libraries to improve model performance on an edge device. A small three-person team delivered the work, provided interim and final reports, and completed the engagement in a short time frame. The client was satisfied with the outcome and the team's project management.

Key Challenges

Improve machine-learning models' performance on an edge device with limited computational power
Increase models' run-time speed

Project Deliverables

Optimized PyTorch machine learning models
Performance-optimization experiments for model run-time
Mid-project and final progress reports

Project Solution

The vendor tested different modeling approaches, PyTorch models, and supporting libraries to optimize machine-learning models to run faster on a constrained edge device. They ran experiments, compared schemes and models, and produced interim and final reports documenting progress.

Project Outcome

Client reported successful collaboration and was happy with the outcome (9/10)
Delivered high-quality results despite higher cost
Project management was strong with regular communication and reporting
Completed in a short amount of time

Platforms

Mobile
AI/ML Platform

Tech Stack

Python

Client Endorsement

Overall Review Rating

4.75

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“Xmartlabs produced excellent deliverables that satisfied the expectations of internal stakeholders. Their organization and communicativeness facilitated seamless and transparent collaboration. A team of experts, they completed tasks efficiently to deliver results within a short timeframe.”
Mehdi Mohseni
CEO
Date: 06/01/2020