All topics
Pick your interview domain
Each topic is a structured roadmap with adaptive checks, dimension scoring, and AI feedback — not just a list of questions.
System Design
Master the art of designing scalable, reliable, and maintainable systems
16 modules
Data Structures & Algorithms
Pattern-first DSA training — learn to recognize which pattern applies before writing a single line of code. India-context problems, complexity analysis, and real interview trade-offs.
18 modules
SQL & Databases
SQL from basics to advanced — joins, CTEs, window functions, indexing, query optimization, and database design for real-world systems.
12 modules
LLD / OOD
Low-level design and object-oriented design problems — class diagrams, machine coding, and design patterns tested at top product companies.
25 problems
LLM Core
The foundational LLM knowledge every AI/LLM/Agentic engineer needs: how LLMs actually work, tokens and context, prompt engineering, structured outputs, the provider landscape (OpenAI/Anthropic/Google/open-source), cost economics, evaluation, and production readiness. Reused across LLM Engineer, Agentic AI, AI Engineer, and FDE roadmaps.
22 modules
RAG & Retrieval Systems
The complete Retrieval-Augmented Generation playbook — from chunking and vector databases to hybrid retrieval, reranking, agentic RAG, evaluation, failure modes, and production system design. For engineers building search-grounded AI features that actually work at scale.
25 modules
GenAI Engineering
The complete interview-prep track for AI Engineer roles in 2026. Covers how LLMs actually work, RAG system design, context engineering, evaluation strategy, fine-tuning decisions, multimodal systems, safety and guardrails, production operations, cost/latency tradeoffs, and GenAI-specific system design patterns. Builds the depth and vocabulary to ace both technical and system-design rounds at companies shipping LLM-powered products.
20 modules
Agentic AI
Build and evaluate LLM agents, tool-calling loops, multi-agent systems, and evaluation frameworks — the fastest-growing interview track.
44 problems
Evals Engineering
The definitive evaluation engineering track: from metric theory to production eval platforms. Covers benchmark design, LLM-as-judge systems, RAGAS/DeepEval frameworks, human preference pipelines, regression CI for AI, safety red-teaming, agentic evals, RAG evaluation, multimodal eval, synthetic data generation, production monitoring, and building internal eval infrastructure. For engineers who ship AI systems that actually work.
15 modules
AI Inference & Serving
The complete inference engineering track: from transformer memory math to production-grade multi-GPU serving. Covers KV cache internals, continuous batching, PagedAttention, quantization (GPTQ/AWQ/GGUF), speculative decoding, vLLM/TensorRT-LLM/TGI architecture, model and pipeline parallelism, edge inference, embedding serving, inference hardware (H100/B100), observability, and cost optimization. For engineers building or operating LLM serving infrastructure.
18 modules
Multimodal AI Engineering
End-to-end multimodal AI engineering: from VLM architecture internals to production document AI pipelines. Covers CLIP/SigLIP vision encoders, LLaVA/InternVL/Gemini VLM architectures, diffusion model math (DDPM/DDIM/DiT), ControlNet, Flux, text-to-video, Whisper/TTS audio systems, multimodal RAG (ColPali), VQA eval, multimodal fine-tuning, document AI, video understanding, multimodal agents, safety, and system design at scale. For engineers building the next generation of perception-aware AI products.
18 modules
MLOps
Build, deploy, and operate production ML systems — from data pipelines to model monitoring
20 modules
Data Engineering Core
The DE-specific intellectual core: data modelling, warehousing, Spark, ETL, streaming, orchestration, quality, and governance. Pairs with the SQL deep-dive for the full Data Engineer roadmap.
19 modules
PySpark & Data Engineering
Spark internals, optimisation, streaming, and end-to-end pipeline design for data engineering and ML infrastructure roles.
19 modules