on
4 minutes read
The Man In The Ivory Tower
It’s 11:30pm and Ola is still at his desk at home putting finishing touches to the demo he will be delivering to the data team the next day. This is his first demo he will be giving since he joined this manufacturing company as a Data Scientist I straight out of grad school. He has been preparing for this since his manager told him 6 weeks ago and his manager also said the Vice President of Data might drop in.
Earlier in the day at work he spent some time with Anne doing a rehearsal with her. And since he got home that evening, he has been implementing his tweaks based on Anne’s recommendation. With every click of the mouse that pierces the silence of night, he thinks “WWAD” - What Will Anne Do?
He remembers the first day he met Anne, his mentor. It was a few weeks after he resumed at the company and she introduced herself as a Staff Data Scientist and explained her strategic role and responsibilities that cut enterprise-wide. He had never heard of that title and it sounded so cool bordering on exclusive that someone that high up doesn’t have people-management responsibilities. He couldn’t wait for his next 1-on-1 to ask his manager more questions about this career path and how he can be like Anne someday. In the mean time, he needs to deliver a demo in a couple hours on how to use transformers for speech-to-text tasks. Ola decides to delicately put his laptop to rest to catch a few hours of sleep and calm his nerves.
A rousing round of applause energises his sleep-deprived body as he concludes his demo by answering the last question thrown at him. The VP of Data says a few words which he barely attended to because his mind was somewhere else. He was thinking of some other applications and extensions of what he just demonstrated - “perhaps, I could build an app that transcribes court proceedings as they happen or an app that writes prescriptions as the patient is talking”. Lot’s of thoughts going through his head but he was brought back into the room on hearing his manager’s voice telling him to check in his code into the demo repository - a repository that contains past demos which are all collecting dust and never made it to production. After he checks it in, he is to move on to his next task on the project board.
The task is not about speech-to-text algorithms. Rather it is about delivering a dataset to a product line manager around the uptimes of a particular conveyor belt on the assembly line. The request is familiar because he has written a query in the past to answer a similar question. He doesn’t know what the manager needs it for, neither does he know what products the specific assembly line produces nor the decisions these assembly line managers make with his datasets. He just knocks these requests out of the way and he considers them a distraction because they do not require him to use the latest algorithms or techniques. His career goal is to become a Staff Data Scientist like Anne and he would rather focus on implementing complex algorithms than solve “simple” business problems he thinks.
Data professionals like Ola are more likely to be seen in a centralised data organisational structure. In this type of structure, everything data sits under one organisational unit and engagement with other business units is often in a support capacity. This type of structure often leads to higher job satisfaction because there are various paths for career growth, good mentorship and a strong community of practice within the data organisation. On the other hand, if not managed properly, it can lead to members of the data organisation spending substantial time on things that are not relevant to the bottom line of the business and disconnected from business operations by living in their ivory tower.