Jul 26, 2022
Arcanum supporting ike GPS on their machine learning journey
Background
ike GPS are providing industry leading AI solutions for those provisioning power and communications networks with their pole management platform technology. They had a pretrained model and needed a solution to inference it while supporting auto-scale.
Journey
- Arcanum provided a micro service that infers pole detection for IKE gps, by helping with productionising & optimising their current model solution.
- We used our spellcaster to generate the code, post it to the Bitbucket to create the CI/CD pipeline and helped them provision the AWS infrastructure.
- Guided by the metrics from our MLOps monitoring system, we found their model would benefit from a GPU, resulting in a 1000% performance increase.
- We built them an auto-scale feature to ensure they could continue to process quickly as their demand grew, while keeping costs down during quiet periods.
Business Value
“Overall it’s been a pleasure working with Arcanum — We are in an interesting position of having extensive in-house AI experience, but not a lot of experience with running AI workloads on the Cloud. Arcanum have helped us to gain a better understanding of how we can best move our workloads to the cloud and have successfully implemented a solution for us on AWS that is performing very well.
This service involves using AI to detect whether a particular image in our database contains a pole, and the laser in our device is hitting the pole for that image — If so, the image is billable, otherwise not billable. The primary internal customer for this is our accounting team, who use the information to generate invoices for the customer and doing this in a timely manner at the start of each month is crucial. I was speaking to a member of the finance team yesterday, and she told me that the overall process ran flawlessly this month.
Also, this month, no one has asked me any questions about the process, which is great since I can invest my time in other new innovations without having to be involved in maintaining the existing system.
The main objective was to have Arcanum help us with deploying our existing models into a cloud environment in a Cost Effective, Secure, Scalable and Reliable way. This has been achieved for the solution mentioned earlier, and I am looking forward to building on this foundation for future work. My overall objective is to set up an environment where the IKE team can focus on the things that we are experts at (using AI to help people measure and assess Utility poles so that we can help our customers provide power and WiFi safely and reliably to their customers), and have Arcanum support us with the underlying MLOps and Devops for this.
Arcanum have been extremely responsive in understanding and resolving any issues that arose after the initial deployment.” Jeremy Gold, Principal AI/ML Engineer / Mad Inventor, ike GPS
“I’m stoked by IKE GPS’s passion of moving machine learning model from on-premise to AWS cloud, especially from Jeremy. IKE GPS conveys their requirements to us very clearly, which is to help them deploy their computer vision models with proper CI/CD to AWS and leverage AWS scalable GPU computing. The collaboration from IKE GPS is awesome, which helps the job finish on time. Really appreciate working with IKE GPS and looking forward to working with them again in the near future.” Bin Wang, Head of Engineering, Arcanum.
