My path to Machine Learning at Cleo

Making things work

I love the word 'engineer'. I've always had the greatest admiration for this discipline, I thought scientists come up with theories, sure, but engineers are the ones who make stuff work.

Yet for a very long time it didn't occur to me that the word could ever apply to me. I loved maths yes, but then I loved everything, really, and at school already being good at maths was very much touted as a boy thing.

I still attended after hours advanced maths classes for fun (even though they weren't my official electives, i.e., ancient Greek and history -true story). But my quantitative proclivities waited many more years to be satisfied.

Paths to maths and joining Cleo

My return to maths was a very roundabout, slow process. I did a year of business school, then completed a Bachelor in French and Spanish literature, then a Masters in linguistics and finally PhD in psycholinguistics.

Moving towards linguistics and psychology came with more objectivity and learning about statistical methods and programming, however I did not forget the love of language that brought me to the field in the first place. Discovering NLP was a pivotal point: my two loves, languages and maths coming together!

After My PhD, I ping-ponged between industry and academia for a while before being swooped up by the fierce, fast moving wave that is Cleo.

Cleo-ing in style

When I first started at Cleo, I was expecting my first child. Some companies would have been put off by this, but not Cleo. They welcomed me warmly and were incredibly generous and supportive. It's not easy combining parenting with fast paced fields like tech and AI, but Cleo gives you the space and support you need to thrive.

Cleo's was one of the best onboarding experiences I've had. While there are many things to get up to speed with (being a fast growing company constantly striving to upgrade user experience), you are given appropriate time and mentorship to find your flow.

I would say for an ML engineer, Cleo strikes the perfect balance between structure and independence. There are robust frameworks in place to ensure that work is aligned with company and business goals, milestones are explicitly set and regular communication is scheduled to maximise visibility. However, Cleo is not a company set in her ways or unwilling to listen below senior leadership. Ways of working and product initiatives are constantly assessed and promptly revised if found lacking, creating a cycle of quick, positive iteration and continuous improvement.

MLEs also own their work end-to-end at Cleo. That means they are responsible for bringing projects all the way from idea to production and in-app use, and these are carried through collaboration and brainstorming within 'squads', i.e., topic units where a team of product analysts, frontend engineers, backend engineers, content designers and ML engineers come together under a product manager to deliver features. This guarantees that there isn't a siloing of data science work where half-baked ideas are handed over without follow-up. However, Cleo benefits from a robust ops pipeline and team who are always happy to provide help.

Cleo's culture is one of inclusivity (the true, non token kind) and openness. The latter is among other things manifested in their complete transparency regarding salary bands and criteria for advancement. The former is experienced every day when looking around and witnessing a high diversity of gender, ethnicity and backgrounds. I don't feel out of place because I am a linguist. On the contrary, I feel like my perspective, like that of everyone in the company, is valued, and its out-of-the-boxness, embraced.