Platform Engineering at Personal Capital

Q&A with Donya Izadi

At Personal Capital we get behind being upfront and we love shining a spotlight on our team’s hard work! With our Q&A series, you’ll hear directly from Personal Capital engineers about their favorite projects, how they’re innovating, and how their team works together to create the best experience for our users and clients. In this post, we talked to Donya Izadi about her role in platform engineering.

What is your role at Personal Capital, Donya?

I’m a Senior Software Engineer, primarily working on platform and security projects. I’ve been at Personal Capital for three years.

What does platform and security include?

Primarily this means working on proof of concept or “POC” projects, as well as resolving security issues.

This could require evaluating new ways of hosting the web services that back the Personal Capital app or testing out software frameworks that could solve tricky problems for us.

We have to solve for lots of problems as we deal with critical information about our customers’ personal financial lives.

What does your day-to-day look like?

Our backend services are primarily built with Java, so I spend a fair amount of time writing application code in Java.

Otherwise, I spend a lot of time communicating with our DevOps and Security teams.

What technologies are you excited about that will help you do your job?

I’m excited about using AWS’s Lambda + Kinesis services. They allow for easily handling large volumes of event processing in real-time.

Lambda functions are key to managing a data collection pipeline — it can consume the output of a Kinesis stream, processing the data and persisting the results for other downstream jobs.

I’ve gotten to learn a great deal about optimizing the performance of Lambda functions with Java, Node.js, Python — allocating the right amount of hardware memory and CPU is critical for performance and cost management.

Size of the deployment package is also important, because of an unfortunate property of Lambdas: the “cold start.” This refers to latency experienced when you trigger a function for the first time — it becomes a critical issue when your Lambda is involved in a customer-facing process.

Thanks for joining us Donya!