Skip to main content

Lyearn

My journey at Lyearn was marked by significant contributions across various domains, from backend optimization to infrastructure management and AI implementation.

Product Engineer​

GET /v1/workExperience?company=lyearn
{
	"role": "Product Engineer",
	"startDate": "01-01-2021",
	"endDate": "01-04-2022",
	"skills": [
		"GraphQL",
		"MongoDB",
		"Redis",
		"Kafka",
		"gRPC",
		"NodeJS",
		"TypeScript"
	],
	"status": 200
}

Key Highlights​

When I joined Lyearn, one of our primary challenges was system performance. I took on the task of elevating our backend system, achieving almost a 30% improvement in overall latency of core APIs. This wasn't just about speed; it was about setting best software design practices for the team, creating a foundation for scalable growth.


As our system grew, so did our need for robust infrastructure. I pioneered the establishment of core infrastructure components, ensuring seamless running and easy monitoring of micro-services on Kubernetes. This infrastructure cycle included implementing comprehensive metrics and monitoring systems, giving us very detailed visibility into our application's performance.

Integrations Module​


lyearn-integration


To expand Lyearn's capabilities, I designed and developed a generic integrations module with extensible APIs. This module expanded our third-party application integration efficiency by 50%, opening up new possibilities for our platform.

Product Engineer II​

GET /v2/workExperience?company=lyearn
{
	"role": "Product Engineer II",
	"startDate": "01-04-2022",
	"endDate": "01-04-2023",
	"skills": [
		"REST",
		"PostgreSQL",
		"Elasticsearch",
		"Kubernetes",
		"Terraform",
		"Helm",
		"ELK Stack",
		"Prometheus Monitoring Stack"
	],
	"status": 200
}

Key Highlights​

Reporting Engine​


lyearn-reporting


One of the most impactful projects was the development of a dynamic yet generic reporting module. This module could process tens of millions of documents in parallel, providing insightful data using system-generated metrics. It was a game-changer for our data analysis capabilities. You can find the details of the implementation in this Medium blog.

BOYS (Bring Your Own SAML)​


lyearn-saml


Security was always a top priority. I implemented a generic SAML (Security Assertion Markup Language) solution, enhancing our authentication processes and making it easier for enterprise clients to integrate with our system.

Senior Product Engineer​

GET /v3/workExperience?company=lyearn
{
	"role": "Senior Product Engineer",
	"startDate": "01-04-2023",
	"endDate": "15-01-2024",
	"skills": [
		"Federated GraphQL",
		"Golang"
	],
	"status": 200
}

Key Highlights​

Optimizing for Scale and Performance​

As Lyearn grew, we faced new challenges in scalability and performance. I led several key initiatives to address these:

  1. Tenant Creation Optimization: We achieved a remarkable feat of reducing tenant creation time for lyearn.com to just 1-2 seconds down from a manual effort of about 30 minutes, significantly improving our onboarding process.
info

Well the best way to experience it is by trying it out 😉 lyearn.com.

  1. Database and Search Optimization: I maintained authoritative expertise on Elasticsearch, optimizing query performance and developing custom Elasticsearch plugins. This resulted in a 37% improvement in query response time. I also stabilized Elasticsearch by adding a distributed locking gatekeeper, adding a node level cache and partitions for better performance.

lyearn-saml


  1. Load Testing: I conducted extensive load tests, stress tests and SOAK tests, ensuring our system could handle high-stress scenarios.

Architectural Transformation​

As our system complexity grew, I orchestrated a strategic transition to Federated GraphQL architecture from a single GraphQL server. This move enhanced service scalability and promoted separation of concerns, allowing our team to work more efficiently on different parts of the system.

Technology Migration​

I identified critical sections in core API code which were CPU intensive and creating a cascading effect of increased waiting times on incoming request in NodeJS and executed a migration to Golang. This transition resulted in an impressive 80% reduction in processing time, showcasing the importance of choosing the right tool.