🧠 AI System Design Blueprints
LLM serving, RAG, vector search, training infra, agents, evals and cost control — designed from first principles with real numbers.
3 Tracks
22 Published
0 Free
22 Premium
100 Planned
🧠 Track 1 10 posts
LLM Serving & Inference Infrastructure
10 of 10 published
- 01 Build an LLM Inference Gateway with Load Balancing Across Model Replicas Up next ★ Premium ★
- 02 Build a Token Budget Management System for LLM APIs Up next ★ Premium ★
- 03 Build a KV-Cache Sharing Layer for Multi-Turn LLM Conversations Up next ★ Premium ★
- 04 Build a Speculative Decoding Engine to Cut LLM Latency by 3x Up next ★ Premium ★
- 05 Build a Continuous Batching Server for High-Throughput LLM Inference Up next ★ Premium ★
- 06 Build a Disaggregated Prefill-Decode Inference Architecture Up next ★ Premium ★
- 07 Build a Model Sharding System for Serving 70B+ Parameter Models Up next ★ Premium ★
- 08 Build an LLM Cold Start Optimizer for Serverless Inference Up next ★ Premium ★
- 09 Build a Prompt Compression Pipeline to Reduce Token Costs Up next ★ Premium ★
- 10 Build a Multi-Modal Inference Server for Text, Image, and Audio Up next ★ Premium ★
🔍 Track 2 10 posts
RAG & Knowledge Systems
10 of 10 published
- 01 Build a Production RAG Pipeline with Hybrid BM25 + Dense Retrieval Up next ★ Premium ★
- 02 Build a Chunking and Embedding Pipeline for 100M Documents Up next ★ Premium ★
- 03 Build a Knowledge Graph-Augmented RAG System Up next ★ Premium ★
- 04 Build a Multi-Hop RAG System for Complex Question Answering Up next ★ Premium ★
- 05 Build a Real-Time RAG Pipeline with Streaming Document Ingestion Up next ★ Premium ★
- 06 Build a Private Enterprise RAG System with Row-Level Security Up next ★ Premium ★
- 07 Build a Citation Tracing System for RAG Responses Up next ★ Premium ★
- 08 Build a Document Freshness and Re-Embedding Pipeline Up next ★ Premium ★
- 09 Build a Conversational Memory Store for Long-Running RAG Sessions Up next ★ Premium ★
- 10 Build a Query Rewriting and Expansion Engine for RAG Up next ★ Premium ★
🗃️ Track 3 2 posts
Vector Databases & Embeddings
2 of 10 published