Explore AI engineering
Build your understanding of AI, retrieval, data platforms and distributed systems. Pick a topic and work through it at your own pace.
Advanced AI Interview Lab
Work through RAG failures, enterprise retrieval, KV-cache capacity, large-scale serving, GPT-style training, post-training and reasoning budgets. Eight worked lessons with interview answers and self-checks.
RAG, LLM Systems & GPT Design
Reference designs, explicit assumptions, capacity calculations and failure analysis. Sources reviewed 16 October 2026.
Advanced RAG: retrieval, reranking and grounded answers
A relevant-looking passage is not necessarily enough to answer a question. Diagnose failures at ingestion, retrieval, ranking and generation before changing the model.
Enterprise RAG: permissions, freshness and index migrations
Design retrieval that respects changing permissions and document versions. A good answer delivered to the wrong person is a system failure.
LLM serving: KV cache, GQA and memory budgets
Estimate what fits on an accelerator before promising a concurrency target. Separate model weights, KV cache, activations and runtime reserve.
LLM serving: prefill, decode and speculative sampling
Optimize the stage that limits the user experience. An improvement in aggregate tokens per second can still make an interactive service feel slower.
Design a GPT-style decoder and training pipeline
Start from tokenization and a causal objective, then work through training memory, distributed execution and evaluation. This is a reference decoder design, not a claim about private GPT internals.
Post-training, preference learning and release evaluation
Choose the intervention that matches the failure. More supervised data, a preference objective and a better evaluation set address different problems.
Frontier systems: MoE, reasoning budgets and tool execution
Distinguish sparse computation, longer inference and tool access. Each changes a different resource or capability and introduces a different failure mode.
Design a large-scale GPT-style chat service
Turn model inference into a reliable product: conversations, streams, admission control, regional failure handling and a release process that protects users.
JAX for ML Engineering
Build differentiable, accelerated numerical programs with JAX and measure their performance correctly.
Transformations & Training
From array operations to a compiled gradient step.
Jev: Typed Decisions for Agents
Use TypeSafe AI’s System One model for classification, routing and rubric-based assessment, with measured decision thresholds.
Decision Design & Evaluation
Design a decision gate and evaluate it against human labels.
Agentic AI & Multi-Agent Systems
Master autonomous agent architectures, LangGraph stategraphs, Model Context Protocol (MCP), human-in-the-loop, and multi-agent coordination.
Agent Foundations & State Machines
Deconstructing workflows vs agents, statefulness, reasoning loops, and deterministic graph execution.
Workflows vs Autonomous Agents & ReAct Loops
A software workflow follows a predetermined path through deterministic tasks and LLM calls. An autonomous agent, in contrast, actively determines its execution route at runtime using an iterative cycle of Observation, Thought, and Action.
Model Context Protocol (MCP) Architecture
Model Context Protocol (MCP) is an open standard that decouples AI models from proprietary tool integrations, enabling secure, modular connections between LLM applications and data sources.
RAG University: Production Retrieval Systems
From Naive RAG to Hybrid Search, Cross-Encoder Reranking, Graph RAG, and Parent-Child Chunking at enterprise scale.
Chunking, Embedding & Hybrid Retrieval
Designing the physical retrieval pipeline: tokenization, dense vs sparse search, and reciprocal rank fusion.
Advanced Chunking Strategies: Semantic, Hierarchical & Late Chunking
The greatest determinant of RAG accuracy is not the LLM or the vector database: it is chunking quality. Bad chunking breaks ideas mid-sentence, splits tables, and destroys semantic coherence.
Hybrid Search & Cross-Encoder Reranking
Vector search alone struggles with acronyms, part numbers, and exact keywords. Sparse search (BM25) struggles with synonyms and intent. Hybrid search + Reranking gives you the best of both worlds.
LLM Engineering & Core Transformers
Deep dive into Transformer Attention, KV Caching, LoRA/QLoRA Parameter-Efficient Fine-Tuning, Quantization, and Speculative Decoding.
Transformer Internals & Inference Acceleration
The mathematical engine: Self-Attention, Key-Value cache optimization, and quantization schemes.
Distributed Big Data Engineering
Petabyte-scale Apache Spark tuning, Adaptive Query Execution (AQE), Apache Kafka streaming, and Apache Iceberg lakehouses.
Spark Internals & Skew Optimization
Catalyst optimizer, shuffle partitions, AQE, and broadcast thresholds.
Distributed & AI System Design
Enterprise RAG Platforms, Real-time Fraud Detection, Autonomous Coding Agents, Distributed Rate Limiters, and Kafka Event Sourcing.
Enterprise AI & Scalable Platforms
Designing end-to-end multi-tenant AI systems with strict SLAs, guardrails, and cost controls.