TakingTheLeapFromFinanceToTech

I’m not usually someone who likes to talk about themselves, but I often get asked how I made the move from finance to development. From the outside it might seem like a huge shift, and in some respects it was. But in reality, it was a gradual process that took years of self-learning to be comfortable taking the leap.

Finance

After university, I decided to start my career in finance as I was really Interested in trading and financial markets. I held several positions in my career, and at each stage I wanted to automate manual processes, which got me deeper and deeper into the coding world.

In the early days, I was working as a Portfolio Manager, responsible for managing all aspects of a client’s portfolio from trading to payments. Every morning I would get to my desk and check to see what actions were required. This was a very manual task, which lead me to my first step towards coding… Excel.

I wanted to figure out a way to analyse all portfolios and cash balances. I also wanted to automate generating reports for client meetings to provide useful data relevant for that client. In the grand scheme of things, these weren’t overly complex tasks, but they did save a huge amount of time for me and the team. They also got me thinking about what other solutions I could create.

My role then moved into investment research, analysing market data and meeting with fund managers to understand their process and view on the world. This is where my ideas really started to ramp up. After meeting with several managers, I began to think about my own investment processes.

Python

I started reading investment research papers and began to learn about things like factor investing after hearing Fama & French mentioned by several fund managers. After trial and error with Excel, I realised I was going to need something more powerful, which is when I moved into Python.

I quickly found out that learning to code by yourself is a very hard process, as you don’t know what you don’t know. So you definitely need some form of roadmap. There are several options, use free resources like FreeCodeCamp or opt for paid content like online courses and bootcamps. I decided to do both, first learning the basics with FreeCodeCamp and then buying several courses from Udemy and also reading various books.

The trouble I found is that learning from courses can be fairly mundane, particularly in the beginning. However, there are some great resources online. DataCamp was a great way to learn how to use Numpy and Pandas, which are two fundamental packages for anything data science related. FreeCodeCamp was a great way to learn more of the basics. And Udemy gave a more in-depth understanding of the language. If I was doing this again, I’d also suggest looking at Leetcode, which allows you to practice solving a variety of small problems in the most efficient way.

So how did I continue to stay motivated? I decided to think of a project that meant something to me. For me, this was to create a portfolio of multiple investment strategies. This meant learning to code, thinking about how a backtesting engine would work and also factoring in how the portfolio should be constructed and trades executed.

With such a large task, I knew I needed to break it down. Fortunately, there was a great platform called Quantopian, which was an open-source hedge fund that provided a free platform for people to build investment strategies. The business model was to run competitions and then for institutional investors to be able to invest in the winning strategies. I wasn’t so interested in entering competitions, but more on using their platform to focus on how to code and my project rather than the infrastructure.

Also there was a great community of people sharing some of their ideas. This meant I could see how other people code. Were there certain patterns they used? Or efficient ways of analysing large sets of data?

Reviewing other people’s code is fundamental throughout your career. It ensures teams come up with the best solutions but it can also be a real eye-opener on how things should be done. This ultimately allowed me to get to a point where I could competently build out a range of different strategies.

Since then, I’ve continued to do all sorts of learning from data structures and system design to machine learning and AI. This has allowed me to continue my interest in automating finance and to even work on automating oil futures strategies for a Commodity Trading Advisor (CTA).

React / NextJS

The problem with Python is it’s very much a backend language. Whilst it’s amazing to be able to solve such complex problems, it isn’t very user-friendly to someone who has no idea about coding. Don’t get me wrong, you can create user interfaces, but it’s not the area it shines the most. So this is when I started to look at web development.

As I already knew Python, I decided to start building with Django. This allowed me to create websites with authentication, APIs, and a front-end. It also gave me an intro into HTML, CSS and Javascript. But it seemed that much more powerful front-ends could be built with React.

I then went back to more courses, learning about everything from Typescript and React / NextJS to Databases, Performance, Security, Testing and Cloud Computing. Becoming a developer means continually learning. Technology is always evolving, which means you have to stay ahead of the curve.

To implement my learning, I built out full-stack data dashboards connecting to both external and internal APIs using a combination of NextJS and FastAPI. NextJS is a great framework for building complex apps. It’s the primary framework we use at Antler for building out SaaS solutions for clients. But I find Python is better suited for data analysis. Ultimately, it’s about choosing the right tools for the job, which is why I decided to use FastAPI for some data intensive API routes.

Taking the leap

Deciding to leave my 8 year career in finance was a tricky decision. I’d spent a huge amount of time and energy in my journey to becoming FCA regulated and I felt like I had done some good work in the process. Throughout my career, I continued to build a variety of tools to make work easier and more efficient helping in all sorts of areas from tax to compliance. It became the thing I felt I was best at. Ultimately, I got to a point where I felt like if I didn’t give full-time software development a shot now, it would become far harder in the future.

Luckily the Antler team were looking to take on someone new to help with the expanding number of projects and it's been amazing to widen my experience across a huge range of solutions. We’ve built all sorts of apps covering risk analysis, web3, sustainability, branding and more for a range of companies from start-ups to household names. Each project comes with its own set of goals and challenges that we look to solve in the most efficient way.

What advice would I give?

If you’re just beginning your journey, it can be very difficult to stay motivated. For me, the way to learn is to have a problem that I want to solve and to try and solve it.

But to caveat, I am a huge advocate of best practice as it ensures a well maintained codebase as your project grows. Don’t just hack away hoping to find a solution. Instead make sure to get a full understanding of best practices. You can find books dedicated to clean code and design patterns that can help. You can learn about Big O notation to assess performance issues. And getting a solid understanding of creating and validating types will make coding feel more structured and secure - like bowling with the barriers up. For Python this means learning Pydantic and for Javascript, move across to Typescript and get to grips with tools like Zod.

If you’re struggling with how you’re currently learning, try another method like reading a book. One of my favourites is The Imposter’s Handbook. It’s a book written by a self-taught Senior Software Engineer who had imposter’s syndrome having not done a Computer Science degree. It gives a real insight into the development of software and grounds the principles you hear about but might not understand.

If you’re thinking about switching careers but scared, ask if you would regret not giving it a shot when you’re older. For me, this question allowed me to make the change.

If you have questions about making the leap or automating your daily work, feel free to reach out on LinkedIn.