I applied online. The process took 2 months. I interviewed at SFL Scientific (Boston, MA) in Jul 2021
Interview
This was by far the most unprofessional and bizarre interview that I had. The interview starts with a screening call with HR, followed by a data challenege. After submitting the challenge, I was invited to give a presentation on the assignment. They extended an offer 10 days after the presentation, and as soon as I told them that I am ready to accept the offer the HR asked for a Zoom meeting to finalize some details with the CTO!! It turned out they wanted to rescind the offer after stringing me along for a month!
Absolutely terrible interview process. Do NOT waste your time with this company. Was given the audio pipeline classification take home assignment (which has already been thoroughly analyzed by a GitHub repo online). The “grader” who looked at my submission was clearly incompetent on basic ML concepts (ex. what is overfit) and blissfully unaware that their take home challenge has already been publicly solved, expecting candidates to replicate the same solution? Also they apparently can’t read Python files unless they are notebooks…
I applied online. The process took 2 months. I interviewed at SFL Scientific
Interview
I’m relieved to see I wasn’t the only one who encountered a very strange interview process with SFL. After a two-part take-home data exercise that demanded ~20 hours of work, plus three rounds of behavioral and technical phone interviews, I was ultimately made an offer ~2 months after applying. That offer was then rescinded two weeks later when the firm opted not to pursue the contract for which they were most immediately hiring, allegedly because they didn’t feel it aligned with the firm’s mission. Why this decision was made only after all the labor of interviewing me is unclear. Moreover, they have continued posting the same position ever since, suggesting they were maybe planning to hire me for just that single contract, contrary to what I was explicitly told in the interview process. Quite a strange experience, all around!
Interview questions [1]
Question 1
Various, including: What was the last academic paper you read? What is the architectural foundation of recent LLMs? One of the take-home exercises required developing a basic ML classification pipeline for audio data (an interesting problem, at least!)
