CASE STUDY

Alpha Sense Logo

X

Transforming Image Processing with Machine Learning

  • Founded in 2013 by David Fattal, Pierre-Emmanuel Evreux, and Zhen Peng
  • Midsize company
  • Clients include mobile phone and automotive suppliers using Leia Lightfield displays
  • Maintains partnerships in the education, retail, and medical industries


OVERVIEW

Leia Inc. is a startup paving the way for dynamic 3D imaging with stylistic possibilities through Lightfield hardware and content services. Lightfield is an emerging visual medium that transforms existing device displays with lighting effects, rich textures, and 3D depth. The technology creates a more engaging visual experience, enabling immersive, lifelike experiences. 

BACKGROUND

As a leading provider of Lightfield displays and devices, the Leia team was eager to find ways to accelerate machine learning setup and experiments. They had already experimented with machine learning environments and the Weights & Biases toolset, but still struggled with an unmet need: as a fast-paced startup, they needed to spend more time on model building and research, and less time grappling with infrastructure. The team sought a flexible and easy-to-use platform with seamless onboarding to fit into their workflow. 

SOLUTION

After extensive market research and testing on possible platform solutions, the team at Leia Inc. decided on Spell as the ideal fit for its best-in-class model lineage tools, intuitive UI, and smooth integrations. In particular, Spell’s close integration with Weights & Biases— their existing tool for deep learning— was particularly useful and enabled an easy transition when adding Spell to their workflow.


Since the Leia team had previous experience with machine learning platforms and using commands to run Cloud-based scripts, their engineers quickly ramped up on the command language without needing extra time for training. Within a week, new engineers up to speed and running commands on the Spell platform. 


Edward Li, a computer vision engineer at Leia, noted “The uptime to bringing on a new engineer is literally less than a week, and I think that’s something that’s a very big pro for your platform.”


“Compared with our previous workflows, Spell has increased our efficiency by roughly 20%, with the potential for greater increase as the product grows.”

Edward Li, Computer Vision Engineer 

RESULTS

Leia is rapidly growing and anticipates using Spell will supercharge their growth because its ease-of-use makes it so quick to onboard new engineers. Since Spell enables them to use their existing Google Cloud resources, it breaks the magic black box stereotype, instead helping the team understand their cloud usage. In turn, Leia is able to manage the resource usage of their clusters with ease, preventing wasted resources or surprises in billing at the end of each month. 

“Given that it's using our own Google Cloud resources, it helps us understand what’s going on,” said Puneet Kohli, Leia engineering manager. “When we get the bill for Google Cloud, it’s very clear that this was the part for machine learning training jobs and we can map that back to Spell and understand the usage very easily.”


“When we get the bill for Google Cloud it’s very clear that this was the part for machine learning training jobs and we can map that back to Spell and understand the usage very easily.”

Puneet Kohli, Engineering Manager 

Other features making a significant impact on their day-to-day workflow are the GCP storage configuration, defining machine types, and the ability to easily run jobs in parallel. Simple, straightforward UI has made each of these features easy to use. 


Now, the Leia Inc. team has a more efficient way to run experiments, hyperparameter searches, and concurrent jobs, and the team is excited to continue to grow with the Spell platform. 



Streamline Machine Learning Projects with Spell

REQUEST A DEMO

Schedule a demo with a Spell representative to learn how Spell can help streamline machine learning development and increase machine learning ROI.