AI engineering · interview prep
Learning Tracks¶
A production-oriented notebook for AI engineering — model mechanics, agentic systems, infrastructure, evals, and interview-grade depth. Grounded in 2026 job-market data, not theory for its own sake.
Start here¶
-
Interview focus areas
What employers actually test in 2026, ranked by tier. The fastest path from "I know AI" to "I can pass the loop."
-
Reference architecture
End-to-end agentic system design — the canonical blueprint the tracks build toward.
Tracks¶
-
AI Foundation
Model mechanics and the prerequisite concepts — tokens, transformers, training, adaptation.
-
AI Engineering
Agent-first product and system design: RAG, durable workflows, memory, multi-agent coordination.
-
AI Infrastructure
Serving, gateways, inference, vector retrieval, platform ops, cost and performance economics.
-
AI Security & Safety
Prompt injection, guardrail design, privacy, governance, and red-teaming.
-
AI Product Evals
Evals, golden sets, metrics, telemetry, experimentation, and release gates.
-
AI Specializations
Voice, vision, diffusion, multimodal, and domain-specific paths.
-
System Designing
General system design and architecture, scaled up to AI platforms.
-
Coding Exercises
Coding-round practice — from tokenizers and beam search to tool-calling agents.
-
Infrastructure Tooling
Docker, Postgres, Redis, Django, Kafka, AWS — the developer infra under it all.
-
Career
Growth, positioning, and planning for an AI engineering career.
-
Interview Bank
Question banks and answer practice across agents, RAG, coding, production, and system design.