ujjwal rasume brief
Absolutely. For an interview, the important thing is not just memorizing the introduction. If you mention terms like React, FastAPI, RAG, LangChain, FAISS, LLM, REST API, the recruiter may immediately ask, “What is that?” or “How did you use it?”
Your resume specifically mentions these technologies and projects, so I’ll prepare you around those terms.
1. Your Introduction — With Recruiter-Friendly Explanations
React
Simple answer:
React is a JavaScript library used to build interactive and reusable user interfaces. I mainly use it to create frontend components and connect the frontend with backend APIs.
Recruiter may ask:
Q: Why did you use React?
React makes it easier to build reusable components and manage dynamic user interfaces efficiently.
Q: What is a component in React?
A component is a reusable piece of UI, such as a navbar, button, form, or product card.
Python
Simple answer:
Python is a high-level programming language. I mainly use it for backend development, APIs, automation, and Generative AI applications.
Recruiter may ask:
Q: Why Python for AI?
Python has a large ecosystem of AI and machine-learning libraries, and it is also simple and productive for development.
FastAPI
Simple answer:
FastAPI is a Python framework used to build fast and modern REST APIs. I have used it to create backend services and connect them with React applications.
Your resume specifically mentions building backend services and REST APIs using FastAPI.
Recruiter may ask:
Q: What is an API?
API stands for Application Programming Interface. It allows two different software systems to communicate with each other.
Q: What is REST API?
A REST API is an API architecture that uses HTTP methods such as GET, POST, PUT, and DELETE to work with resources.
Q: GET vs POST?
GET is generally used to retrieve data, while POST is generally used to send or create data.
2. Generative AI
What is Generative AI?
Easy answer:
Generative AI is a type of AI that can generate new content such as text, images, code, or other data based on the input given by the user.
Recruiter may ask:
Q: How is Generative AI different from traditional AI?
Traditional AI is often used to classify, predict, or make decisions, whereas Generative AI can create new content based on learned patterns.
Q: Have you worked with Generative AI?
Yes. I have worked with LLM-based applications, RAG pipelines, LangChain, FAISS, and APIs such as Groq and OpenAI. I have also built an AI PDF chatbot that answers questions from uploaded documents.
3. LLM
What is an LLM?
LLM stands for Large Language Model. It is an AI model trained on a large amount of text data that can understand and generate human-like text.
For example, GPT-based models are LLMs.
Recruiter may ask:
Q: How did you use an LLM in your project?
I used an LLM to generate natural-language responses based on information retrieved from documents. In my AI PDF chatbot, the relevant information is retrieved first and then provided to the LLM to generate the final answer.
4. RAG
This is very important because it is directly mentioned in your resume.
What is RAG?
RAG stands for Retrieval-Augmented Generation. It is a technique where we first retrieve relevant information from a knowledge source and then provide that information to an LLM to generate a more context-aware answer.
Very simple example:
Suppose I upload a 100-page PDF.
User asks:
"What is the company's leave policy?"
Instead of asking the LLM to guess the answer:
PDF → Search relevant information → Give it to LLM → Generate answer
That's RAG.
Recruiter may ask:
Q: Why do we need RAG?
RAG helps an LLM answer questions using specific or private information that may not be part of its original training data. It can also help reduce hallucinations by grounding the response in retrieved information.
Q: Explain your RAG project.
I built an AI PDF chatbot where users can upload PDFs and ask questions in natural language. The PDF content is converted into embeddings, stored in a vector index using FAISS, and when the user asks a question, relevant content is retrieved and passed to the LLM to generate a context-aware answer.
5. LangChain
What is LangChain?
LangChain is a framework used to build applications powered by large language models. It provides components for things like prompts, document processing, retrieval, memory, and connecting LLMs with external data.
Recruiter may ask:
Q: Why did you use LangChain?
I used LangChain to simplify the development of my RAG pipeline and to manage the flow between document processing, retrieval, prompts, and the LLM.
Q: Is LangChain an LLM?
No. LangChain is a framework. The LLM is the actual language model that generates the response.
Very important distinction:
LangChain = framework
LLM = AI model
FAISS = vector search/indexing tool
RAG = architecture/technique
6. FAISS
What is FAISS?
FAISS stands for Facebook AI Similarity Search. It is a library used for efficient similarity search over vector embeddings.
Simple example:
Imagine these sentences:
- "How can I apply for leave?"
- "What is the holiday policy?"
- "How do I reset my password?"
Each sentence can be converted into numbers called embeddings.
FAISS helps us find which stored vector is most similar to the user's question.
Recruiter may ask:
Q: Why did you use FAISS?
I used FAISS to efficiently search for the most relevant document chunks based on vector similarity in my RAG-based PDF chatbot.
7. Embeddings
What are embeddings?
Embeddings are numerical representations of text. They convert words, sentences, or documents into vectors so that we can compare their semantic meaning.
Example:
"How do I reset my password?"
and
"I forgot my password. How can I change it?"
Although the words are different, their meanings are similar. Embeddings help identify that similarity.
Recruiter may ask:
Q: Why are embeddings needed in RAG?
Embeddings allow us to represent documents and user queries as vectors so that we can perform semantic similarity search and retrieve relevant information.
8. Vector Database / Vector Search
What is vector search?
Vector search finds information based on semantic similarity rather than just matching exact keywords.
For example:
User asks:
"How can I get my money back?"
Document says:
"Customers can request a refund within 30 days."
Keyword matching may not work perfectly, but vector search can understand that money back and refund are semantically related.
9. REST API
What is REST API?
REST API is a way for applications to communicate with each other over HTTP. It commonly uses methods such as GET, POST, PUT, and DELETE.
Example:
Frontend:
"Give me all products."
React sends:
GET /products
Backend:
Returns product data.
Common recruiter questions:
Q: What are HTTP methods?
GET is used to retrieve data, POST to create data, PUT/PATCH to update data, and DELETE to remove data.
Q: What is JSON?
JSON is a lightweight data format commonly used to exchange data between frontend and backend.
10. JWT Authentication
Your resume mentions JWT-based authentication across your APIs.
What is JWT?
JWT stands for JSON Web Token. It is commonly used for authentication. After successful login, the server generates a token, and the client sends that token with future requests to prove that the user is authenticated.
Recruiter may ask:
Q: Why JWT?
JWT allows the server to authenticate users without maintaining traditional server-side session state for every request.
Simple flow:
Login → Verify credentials → Generate JWT → Client stores token → Client sends token → Server verifies token
11. MySQL
What is MySQL?
MySQL is a relational database management system used to store structured data in tables.
Recruiter:
Q: SQL vs NoSQL?
SQL databases store structured data in tables with predefined relationships, while NoSQL databases such as MongoDB provide more flexible data models.
12. MongoDB
What is MongoDB?
MongoDB is a NoSQL document database. Instead of storing data primarily in rows and columns, it stores data as document-like structures.
Recruiter:
Q: When would you use MongoDB?
MongoDB can be useful when the application needs a flexible schema or when the structure of the data can change frequently.
13. SQLAlchemy
What is SQLAlchemy?
SQLAlchemy is a Python SQL toolkit and ORM. It allows developers to work with relational databases using Python objects and abstractions instead of writing every SQL operation manually.
Your resume lists SQLAlchemy as part of your backend/database stack.
14. Your AI PDF Chatbot — VERY IMPORTANT
This is one project you should be able to explain from start to finish.
Your resume describes it as a RAG-based document assistant using FastAPI, Python, Groq, FAISS, LangChain, and Sentence Transformers.
Recruiter:
"Explain your AI PDF chatbot."
Your answer:
My AI PDF chatbot allows users to upload a PDF and ask questions about its content. First, I extract the text from the PDF and divide it into smaller chunks. Then I convert those chunks into vector embeddings using a sentence-transformer model. These embeddings are stored in FAISS for similarity search. When a user asks a question, I convert the question into an embedding and search FAISS for the most relevant chunks. Those chunks are then provided as context to the LLM through the Groq API, and the LLM generates the final answer. This approach is based on RAG and helps provide answers grounded in the uploaded document.
Interviewer may then ask:
"Why did you split the PDF into chunks?"
Processing the entire document at once can be inefficient and may exceed the model's context limitations. Chunking allows us to retrieve only the relevant parts of the document.
"Why embeddings?"
Embeddings allow us to compare the semantic meaning of the question with the document content.
"Why FAISS?"
FAISS provides efficient similarity search over the embedding vectors.
"Why RAG?"
Because we want the model to answer based on the uploaded document rather than relying only on its pretrained knowledge.
15. AI E-Commerce Project
Your resume describes an AI-powered e-commerce platform built with React, FastAPI, Python, SQLAlchemy, MySQL, JWT and Hugging Face.
Recruiter:
Q: Explain your e-commerce project.
I built a scalable e-commerce platform for small and medium-sized clothing retailers. The frontend was developed using React, while FastAPI and Python handled the backend APIs. MySQL was used for data storage, and SQLAlchemy was used for database interaction. I also implemented JWT authentication and added an AI-powered search chatbot to improve product discovery.
Follow-up:
Q: What was your role?
I worked across the stack, including frontend development, backend API development, database integration, authentication, and AI functionality.
🔥 Most Asked Questions You Should Prepare
For your profile, I would prepare these 20 questions first:
- Tell me about yourself.
- Explain your current role at TnpLab.
- What is React?
- What are React components?
- What are React Hooks?
- What is FastAPI?
- FastAPI vs Flask?
- What is REST API?
- GET vs POST vs PUT vs DELETE?
- What is JWT authentication?
- What is Generative AI?
- What is an LLM?
- What is RAG?
- Explain your RAG project.
- What are embeddings?
- What is FAISS?
- What is LangChain?
- Why did you use LangChain in your project?
- SQL vs NoSQL / MySQL vs MongoDB?
- Explain one project from your resume end-to-end.
Your work experience also specifically involves taking features from requirements through deployment and debugging technical issues, so be ready for questions like “Tell me about a difficult bug you solved” and “How do you take a feature from development to production?”
⭐ One golden rule for your interview
Don't answer like:
"I know React, FastAPI, RAG, LangChain, FAISS..."
Instead, connect everything:
React → frontend
FastAPI → backend/API
MySQL/MongoDB → data storage
JWT → authentication
Embeddings → convert text to vectors
FAISS → search similar vectors
RAG → retrieve relevant information + generate answer
LangChain → orchestrate the LLM workflow
LLM → generate the final response
That single flow will make your GenAI + Full Stack profile much easier for a recruiter to understand.
Comments
Post a Comment