What actually happens in an AI Engineer interview? In this video, you’ll see a real mock interview — not just questions, but how candidates are evaluated in real time. We go beyond answers and focus on: • How to break down AI/ML problems • How interviewers assess your thinking • Follow-up questions that test depth • Common mistakes candidates make If you're preparing for AI Engineer, Machine Learning, or LLM roles, this is exactly the kind of interview experience you need. Watch till the end to understand how top candidates stand out. Follow for More: Youtube : @thinkinmodels X : https://x.com/thinkinmodel Linkedin : www.linkedin.com/in/karthik-varma-94698b56 Instagram: https://www.instagram.com/thinkinmodelslabs Subscribe for more real interview simulations and AI career insights. 00:00 Intro 00:06 Document Processing & AI Accuracy Challenges 02:18 Candidate Background & AI Experience 06:23 NLP: Movie Review Classification 08:32 LSTM vs Transformers for Sentiment Analysis 12:28 Chatbot Design using RAG (Invoices) 14:32 OCR Challenges in Structured Data 18:32 KNN for Prompt Retrieval 20:16 KNN & Embedding Similarity 23:41 Detecting Cheaters in Coding Platforms 25:14 Outlier Detection for Cheating 28:42 Graph Databases for Cheating Source Detection 30:27 Complexity of Cheating Detection in AI 33:48 Data Structures & ML Metrics in Interviews 35:29 Precision vs Recall Explained 38:58 Object Detection & Evaluation Metrics 40:44 YOLO Bounding Box Explanation 44:20 Naming AI Models (Interview Challenge) 46:08 Local vs Cloud Models (Privacy Focus) 49:39 Rise of Lightweight Specialized Models 51:19 AI Agent Coordination Limitations 54:49 Code Analysis Tool: Challenges & Improvements 56:28 Benefits of Open Sourcing Projects 59:42 What Interviewers Really Look For 01:01:17 Fine-Tuning vs RAG 01:04:35 Simpler Systems = Better Accuracy 01:06:16 Importance of Theoretical Knowledge 01:09:23 Conclusion & Wrap-up #AIEngineer #MockInterview #MachineLearning #LLM #TechInterviews
ADVERTISEMENT
Please make more videos ❤
he said BoW it also sparse and has neither semantic nor contextual provider very old technique , if word is present put 1 else zero
and how someone use for such limited data neural archtiecture RNN, or its varients even said transformer .... so if it is text embedd it and bring it into tabular fom use any ML algorithm for classification
the question was not clear more for classification of review , said 500 then it is not real time data ,so it it is simple suprvised problem , we have data set that has features then we are going to use classical mL for classification and there is parameter one versus rest for multiclass , it we have just 2 columns one is for text and other is target then need emedding but word2vec or Glove not provide contexual embedding which will might wrong classification regards
Today I got to know that I know the answers of about 90% of the questions asked in the interview, but the only thing i am lacking in is to articulate my answer precisely as this candidate did. I just forget what i was saying, and sometimes while giving answer to a particular que, i feel like what was the question he just asked, and then i blubbered and fug the interview. Is it only me or somebody else also going thorough the same. And also if somebody has already overcome this situation, I'd really like to hear some genuine tips and tricks to improve this part of mine
very informative to test my knowledge and prepare for future, appreciate it
20:40 KNN typically uses Euclidean distance to identify the nearest data points to a given query point. In contrast, cosine similarity measures the angular difference between vectors rather than the absolute distance between their endpoints. Consider two vectors that have a very small angle between them but differ significantly in magnitude—one is short while the other is long. Using Euclidean distance, the endpoints of these vectors may appear far apart, suggesting low similarity. However, from a directional perspective, the vectors are actually quite similar because they point in nearly the same direction. In such cases, relying solely on Euclidean distance can be misleading. Cosine similarity is often more effective because it focuses on orientation rather than magnitude by computing the cosine of the angle between vectors and the query vector. Of course, in practical systems, cosine similarity is not computed against every vector exhaustively. Various optimization and indexing techniques are used to efficiently retrieve the most relevant vectors.
Bro, in my org, they have provide copilot license to develop software , no like integrating AI in web app. So what should my focus to build a strong profile?
Bhaiya make more such videos...
I don't think we need RAG for the document processing. Second thing, before using OCR, we have to ask what's format of the PDF because OCR is only useful if the PDF is from images, otherwise there are existing libraries that will let you convert a PDF into the text. Once we have the text, either via OCR or the libraries, we can use regex to get information.
for the sentiment analysis in the case of 500 review ill probably use a GenAI model coz If i use a ML model probably need more data to train a model also then tweak the model too much work for 500.
can u also make a video on how to get jobs in ai field , as ai eng jobs are very less compared to sde in india especially for freshers
We can improve attention by using two separate techniques 1. memory Saving technqiues: (comparison still O(n^2) but fast due to less blocks) a) MQA --> multiple query but one key and value blocks b) GQA --> multiple query but grouped key-values blocks 2. reducing time complexity: a) random attention (attending random tokens from a current one and not all) b) sliding window (attending the prev and next tokens only within the respective window) c) there are many variants of sliding window d) Linear Attention (we do Q(Kᵀ V) instead of (QK^T)V)
He has no idea of what he is saying. Just throwing in random technical jargons
Please bring more such videos! Its such a great help to someone like me who is a fresher and stuggling with the job market and interviews right now :)
We want a structured mentoring!!! Are you on topmate or any such platforms (if so plzz keep it affordable).
Very good learning and exposure to how AI/ML interviews are! Thank you very much for making this available for us.
very informative ...Thanks for sharing
please make more similar videos
00:30 , bruh anyone who understands AI/ML knows we don't go to 100 percent accuracy as it'd most certainly meaning extreme overfitting