Hello Blind Community! I just started interviewing for the senior Ml roles and I feel like I am struggling with the system design rounds. Are there any good resources to learn Ml system design interview tips and tricks. I’m already in the Ml industry ideating, building and delivering search retrieval models. I haven’t had much exposure to recommendations and ranking yet ( I consider search retrieval, recommendations and ranking to be quite analogous to one another) Even though I have read blogs about various recommendation and ranking systems of different tech companies, I still struggle to make a good case for myself during system design interviews. Most of the times I struggle to come up with good features to engineer for the task given. And I’m not sure how much to focus on the software engineering aspects of Ml during these rounds. Would appreciate any insights from experienced Ml practitioners here on how to give a perfect Ml system design interview(blogs and YouTube videos alone are not working for me at the moment) TC: 150k #tech #ai #machinelearningengineer
India
Yesterday
555
Who are these retards asking for dictatorship in India?
2024 Presidential Election
Yesterday
544
Heartwarming peaceful protests
Tech Industry
Yesterday
3556
Asians - what are your thoughts on asian female white male ?
Tech Industry
Yesterday
1184
The end of Backdoor Roth?!
Tech Industry
Yesterday
1057
How common is it actually to earn more than 300K TC?
Why is learning old classical ML algorithms even a thing? It's all LLM now which is a whole new industry.
Wait till you use it for prod and see the way they hallucinate.
It’s close to impossible to deploy llms for traditional search and recommendation tasks
I have a similar situation. Had an interview at Meta recently, still waiting for the final decision, but looks like it will be negative because of ML System design interview. Which resources have I used: - This github page is pretty useful. https://github.com/alirezadir/Machine-Learning-Interviews/blob/main/src/MLSD/ml-system-design.md Explains the structure of the answer and gives a lot of useful references. But generally, content is pretty compressed and difficult to understand if you are not in context. - I have found a mentor on mentorcruise.com who helped me with the mock interviews. There are not so many mentors that prepare people for the interview, and I found one mentor with 10 YoE in the industry and good knowledge of ml sys des interviews and how they are estimated. She did a really good job of returning me feedback on my weak points, but she was not so helpful on how to improve them. - Meta hr said that the ml sys interview would cover only classification and recommendation, so I focused on these two areas. - I have found that I forgot some important concepts about ranking that I learned 2 years ago, so I created flash cards (I used the NeuraCache app for that) to repeat and memorize them. Overall, my biggest issue was the clarity of explanation of my approach to the interview because I wasn't sure about it and felt a lot of stress. To overcome that, next time I will also - read and memorize companies' blog posts about their ml solutions (you can find some of them on the first link) - I don't know how, but will make much more mock interviews before the real one. Let's say 20
Tc
Posted
Lol