NLP Engineers are responsible for the interaction between everyday human language and a computer’s ability to process and analyse natural language data. In our latest post we welcome Rita Anjana to the team and learn more about the day to day tasks of a NLP engineer and all you need to know to become one too!
Hi Rita, welcome onboard! Tell us why you were interested in joining Aveni and what your role entails?
I’ve been working as an Applied NLP Engineer for the past few years, primarily within the Fintech/Cybersecurity space. I came across Aveni recently, which seemed to be involved in applying various cutting-edge NLP research to solve extremely challenging problems within this space. After further exploring and talking with the Applied NLP Team behind Aveni, I was convinced this would be the best place to work in, I was specifically impressed with the sharp focus on solving real world challenges and their values as a company.
On a day to day, my work lies somewhere at the center of the research-production engineering spectrum where I’m involved in rapid prototyping of interesting NLP algorithms, training various types of ML models to deploy them and a lot of discussions around these.
What’s it like working for a startup and how does it compare to your previous roles?
Speed at which we build things is a key differentiator between a traditional place and a startup, and for me speed, but with a sense of craftsmanship implies reliability and trust, essentially this means working at a startup you follow the philosophy of “Moving fast, thinking long term.”
I’ve been mostly working at startups which are at various stages of their journey, from early stage to slightly advanced ones for a while now, in that sense this role is quite similar.
What skills do you believe are crucial when working as an NLP Engineer?
If I had to pick one key skill for an NLP Engineer, that would be the ability to understand the Business Problem in the right way, and then framing it as an ML problem and further followed by measuring it with well defined metrics and making sure it’s aligned with the business goals is the primary key. It’s very easy to go down the rabbit hole of trying all kinds of ML/NLP modelling and experiments as an NLP Engineer, but being disciplined about it, guided by the real impact of the systems we design is crucial.
What resources do you find beneficial to upgrade your skills?
I find it quite useful to keep yourself updated on the cutting edge research via NLP subreddit and mostly Twitter, there are also a lot of interesting podcasts like dataskeptic, TWIML podcasts etc.
What is one piece of advice you would give to those looking to enter the field of NLP?
Having a solid foundation in ML Engineering, especially how to build datasets, how to handle various data related issues and a solid grip on Current NLP techniques like transformers is super useful. Along with this the ability to effectively use various standard NLP tools like AllenNLP, spacy, gensim, sklearn, deep learning libraries like Pytorch, Torchtext etc is also valuable.
If you are interested in working for one of the UK’s fastest growing startups, check out our careers page and find out how you can be part of our journey
