project

0

Last Updated:

Freddie Mac

Financial Engineer | Investments and Capital Markets

Post image

Welcome to the FARM

Position

Financial Model Development Professional

Period

August 2023 - Present

Tech Stack

  • C++
  • Python
  • Jenkins
  • Matlab

My first role out of graduate school landed me hundreds of miles away from my beloved Brooklyn, NY. I was hired by Freddie Mac to work under the Investments and Capital Markets division, as a financial model developer on the FARM (Framework for Analytics, Research and Modeling) team. I will probably never get over how badass I thought the team acronym was, like some top-secret facility for training the next generation of spies. In the acronym's defense, the work being done did merit the credit being due. The team was tiny but played a critical role in the company's day to day operations. We were responsible for building out what I can best describe, as being the core calculator that the firm utilized in analyzing their mortgages and portfolios. We weren't the quants designing the models, nor were we the traders who used them. We sat organizationally smack in the middle between the two, acting as the implementers who needed to know and understand both sides. In this role, I picked up the mathematics and data science skillsets the quants used and learned the financial needs of those who sat on the trading floor (albeit on a somewhat novice level).

My day to day revolved around implementing financial models in C++, which where then applied to portfolios of MBS securities across a distributed network of servers. Wanted to learn the prepayment risk of a loan? We had that logic. Wanted to predict house price appreciation? Yup, we got it. Predict how interest rates would affect the portfolio? Of course.

Without going too deep into the sensitive work, I was responsible for helping the firm solve findings from FHFA (Federal Housing Finance Agency), implement new models for house price appreciation (including the implementation of Monte Carlo simulations) and augmenting mass tie-out scripts to be used for thousands of loans to ensure that FARM's implementation was correct to the specifications provided within a tolerance difference of 1.0E-14. The team being detail oriented was an understatement.

For the time being, I will leave the writeup here, due to the confidentiality of the role, systems and processes being done. And because I want to respect all the parties involved and not land in hot water.

I will probably move this content around as I create a separate part for my site to house work experience and the like, but figured it would still be interesting to discuss my experience here as it showcases my work as a software engineer.

How was this article?

If you want to send me specific feedback, please do so on the Contact Page. Thank you!