One place to prep for AI interviews and exams.
A structured platform for AI practice. Every question ships with tiered answers, deep dives, common mistakes and follow up questions. Spaced repetition on your weak spots, instantly evaluated answers, company packs, and formats built for real interviews and assessments.
Topics that interviewers actually ask
Attention Mechanism
The mechanism that lets each token attend to every other token. Self-attention, multi-head, scaled dot-product, MQA, GQA, FlashAttention, KV cache.
Fine-Tuning
Adapting base models for downstream use. SFT, LoRA, QLoRA, PEFT, distillation. When to fine-tune vs RAG vs prompt engineering.
Context Engineering
Managing what the model sees. Context window strategy, memory, compression, lost-in-the-middle, summarization-as-context, RAG-as-context patterns.
Retrieval-Augmented Generation
Grounding LLM responses in external knowledge. Chunking strategies, retrieval, reranking, query rewriting, hybrid search, generation, end-to-end evaluation.
AI Agents
Multi-step LLM systems. ReAct, Plan-and-Execute, memory, tool use, planning, reflection, multi-agent orchestration, trajectory evaluation.
Inference Optimization
Serving LLMs at scale. KV cache, FlashAttention, speculative decoding, quantization, batching, vLLM, TensorRT-LLM, TTFT vs TPOT tradeoffs.
Reasoning Models
Frontier inference-time compute. o1, o3, DeepSeek R1, thinking tokens, process reward models, extended chain-of-thought, MCTS-style search.
LLM Frameworks
Picking and using the right abstractions. LangChain, LlamaIndex, DSPy, AutoGen, CrewAI, Mastra, Vercel AI SDK, when to use each and their tradeoffs.
LLM System Design
End-to-end production architectures. Multi-tenant serving, rate limiting, prompt caching, model routing, cost models, latency budgets, fallback strategies.
Prep for the company you're actually interviewing at
OpenAI
Frontier AI lab building GPT-class models, Codex, and ChatGPT.
Cohere
Canadian enterprise AI lab behind Command R+, Embed v4, and the industry-standard Rerank model, RAG-first by design.
Mistral AI
Paris-based AI lab shipping open-weight and commercial frontier models, Mistral Large 3, Mixtral MoE, Codestral, Pixtral.
Databricks
Data + AI platform behind DBRX and Mosaic AI Agent Framework, owns MosaicML (acquired 2023 for $1.3B) and runs Spark at planetary scale.
Anthropic
AI safety lab behind Claude, focused on building reliable, interpretable, steerable AI systems.
Microsoft
Microsoft AI products (Copilot, Azure AI Studio) and Microsoft Research.
NVIDIA
Hardware + software stack for AI, GPUs, CUDA, TensorRT, NeMo, Triton.
Perplexity
AI-powered search engine, the canonical production RAG product.
Cursor
AI-first code editor; agent-driven coding workflows.
Sarvam AI
Indian AI startup building foundation models for Indian languages.
Big-tech AI employer behind Gemini, Imagen, and large-scale ML infrastructure (Brain + DeepMind research).
Meta
Big-tech AI employer behind Llama, PyTorch, and large-scale recommender + GenAI systems.
9 question formats. Pick one.
The answer is only the start. Deep dives, mistakes, and follow ups interviewers ask.
Personalized daily set
A mix of your weak topics, new questions, and review, picked for you every day. Pro tier.
Instantly evaluated answers
Write or speak your answer and get a score, missing concepts, and an improved version. Instant feedback built for real interview and exam prep.
Mastery tracking
Per topic mastery score updates with every attempt. See where you are; know what to drill next.
Company packs
Curated question sets shaped by interview patterns at top AI labs and AI-native startups.
Spaced repetition
Questions you missed come back on a schedule tuned to how you answered. Pro tunes the schedule per question.
Readiness score
Know how ready you are per role: GenAI Engineer, RAG Engineer, AI Product Engineer, AI Researcher, LLMOps Engineer, and AI Product Manager.