BUSINESS SITUATION
The client engaged us to set up and integrate ML-based solutions into its system.
SGA APPROACH
- We proposed multiple end-to-end architectures with a feasibility study and pricing details to provide options to client targeting to cater to the specific needs of low cost and maintenance alongside high-security standards.
- Apart from access and high-security standards, the major challenge we faced was providing real-time model results for streaming data. To resolve this, we divided the process into chunks of serverless micro-services and event services, building the bridge between need, speed, and security.
- We automated all the workflows via code templates to strictly minimize external efforts and avoid any manual interference in a production environment.
- We simplified heavy technical architectures and handy, detailed documentation to ease business communication.
ENGAGEMENT
We established a bi-weekly sprint framework where we utilized Jira boards provided by the client, which allowed us to effectively manage and prioritize tasks, ensuring the timely completion of deliverables.
BENEFITS & OUTCOME
- We streamlined end-to-end architecture with fault tolerance in place.
- We implemented the recommendation engine for new and existing customers by increasing potential customer experience.
- We reduced efforts for the client’s Engineering team.
KEY TAKEAWAYS
- With buzz for recommendation engines, recommendations for investment options is the need of the hour. Personalized investment recommendations cater to audience experience leading to higher user retention.
- Rather than a monolithic approach, micro-services provide better reliability to incorporate versioning changes.
- Financial data needs to be highly secured; following high-security standards and encrypting sensitive data are the keys.
