Kits by AI

Home / Guides

Free guides

Build an AI project you can actually explain

How RAG works, how to build one on free tools, what interviewers ask, and project ideas your examiner will respect. Written by an engineer, not a content farm.

Start here

RAG project for students: how it works, 10 ideas and source code

A RAG project is an app that finds the relevant passages in your documents first and then asks an LLM to answer only from them, with citations. It is one of the best final-year and resume projects in 2026 because it is small enough to finish and deep enough to defend in a viva or interview.

Read the guide · 11 min →
p. 21
RAG

How to build a RAG app on free tools, step by step

How to build a RAG application step by step with Next.js, Gemini and Supabase pgvector: real SQL and TypeScript code, common errors and free deploy.

6 min readRead →
QA
Interviews

RAG interview questions: 30 questions with model answers

30 RAG interview questions and answers: chunking, embeddings, vector databases, hallucinations, RAGAS evaluation and system design, with model answers.

15 min readRead →
?
Interviews

Generative AI & LLM interview questions, answered

30 LLM interview questions and generative AI interview questions with model answers: transformers, fine-tuning, LoRA, RLHF, prompting, agents and safety.

13 min readRead →
Project ideas

Final year projects for CSE: 40 ideas, plus how to defend one

40 final year projects for CSE with difficulty, stack and the twist that impresses examiners, plus a 12-week plan, report outline and viva questions.

16 min readRead →
Project ideas

AI projects for students: 25 ideas from beginner to GenAI

25 AI projects for students, from beginner ML to GenAI and LangChain, with free datasets, APIs and tools for each, plus how to make one resume-worthy.

11 min readRead →
$ python rag.py
Free code

RAG project starter: chat with a PDF in one Python file

The whole RAG pipeline in about 100 lines of Python. Runs on the free Gemini API with no database, so you can see it work on your own notes in minutes.

GitHub · MIT licenceOpen repo →

New to RAG? Read them in this order

Three guides take you from "what is RAG" to a project you can defend in an interview.

  1. Understand it

    What RAG is, how a real app is put together, and 10 project ideas.

  2. Build it

    Step-by-step on free tools: Gemini, Supabase pgvector and Vercel.

  3. Defend it

    30 interview questions with the answers a strong candidate gives.