Tag: #llm
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 208 posts
Context Engineering — How to Design Memory for AI Agents
The era of prompt engineering is fading, and context engineering has emerged as the core skill. This post covers the economics of the context window, memory tier design, patterns for automatically extracting facts from c
2026-06-12 · 17 min read #ai#context-engineering#llm#ai-agent#memoryTop LLM Papers 2024-2026 - Llama, DeepSeek, Qwen, Mistral, Phi, RLHF, DPO, CoT, RAG, FlashAttention, vLLM Reading List
A curated reading list of 30+ must-read LLM papers for engineers building with LLMs in 2024-2026. Covers foundation models (Llama 3/4, DeepSeek-V3/R1, Qwen3, Mistral, Phi-4, Gemma 3), training innovations (MoE, MLA, GQA)
2026-05-16 · 19 min read #llm#papers#llama#deepseek#qwenLLM Fine-tuning Frameworks 2026 — A Deep Dive into Axolotl, Unsloth, LLaMA-Factory, TRL, PEFT, and TorchTune
A complete map of the 2026 LLM fine-tuning ecosystem. Open-source frameworks like Axolotl, Unsloth, LLaMA-Factory, TRL, PEFT, and TorchTune. LLM Foundry (MosaicML, acquired by Databricks). Cloud fine-tuning APIs from Mod
2026-05-16 · 28 min read #llm#finetuning#axolotl#unsloth#llama-factoryLLM Serving & Local Inference in 2026 — vLLM / llama.cpp / MLX / Ollama / LM Studio / SGLang / TGI Deep Dive
A map of the 2026 LLM serving and inference landscape. Datacenter camp (vLLM, SGLang, TGI, Triton, TensorRT-LLM), local camp (llama.cpp, MLX, llamafile, Ollama, LM Studio, GPT4All), emerging camp (KTransformers, MLC LLM,
2026-05-16 · 25 min read #llm#model-serving#inference#vllm#llama-cppAI Agent Frameworks 2026 Deep Dive — LangGraph, AutoGen, CrewAI, Semantic Kernel, OpenAI Agents SDK, Claude Agent SDK, Pydantic AI
A May 2026 deep dive into the ten frameworks that dominate production agent systems — LangGraph 0.3, AutoGen 0.4, CrewAI Flows, Semantic Kernel Agent Framework, OpenAI Agents SDK, Claude Agent SDK, Pydantic AI, Smol Agen
2026-05-16 · 17 min read #ai-agent#langgraph#autogen#crewai#semantic-kernelBuilding an AI Teammate with a Slack Bot — Wiring Up Claude, Gemini, OpenClaw + Extending Tools with MCP (2026 Hands-On)
The highest-leverage surface for deploying AI isn't the IDE — it's Slack, where the whole team uses it, right where the work happens. This is a follow-along hands-on. Create a Slack app, stand up a minimal bot that respo
2026-05-14 · 16 min read #slack-bot#llm#claude#gemini#openclawReasoning Models in 2026 — A Deep Dive on o3, o4, DeepSeek R1, Claude Thinking, Gemini Deep Think, and QwQ
It has been about a year and a half since o1 (Sept 2024) opened the test-time compute axis. In 2026, 'reasoning models' are no longer a separate family — they are a mode that every frontier model can enter. This guide la
2026-05-14 · 20 min read #reasoning-models#o3#o4#deepseek-r1#claude-thinkingLLM Landmark Papers Guide — From Attention to GPT, LLaMA, DeepSeek, o1, and Claude (with References, 2026)
Where do the real shifts in LLMs come from? From Attention is All You Need in 2017 to the reasoning models of 2026, this guide organizes the 20-odd landmark papers you must know, by era and theme. Each paper is compresse
2026-05-14 · 15 min read #llm#research-papers#transformer#gpt#llamaAnatomy of an AI Harness — The Scaffolding That Turns a Model Into an Agent (Loop, Tools, Context, and Building Your Own)
The model is the engine; the harness is the car. Put a different car around the same engine and the driving experience changes completely. In 2026, the real center of gravity in AI engineering isn't the model — it's the
2026-05-14 · 16 min read #ai-harness#ai-agent#llm#agent-loop#tool-executionAI Engineering in Practice — LLM API, RAG, Agents, LoRA/DPO, Vector DB, Evaluation, Observability, Prompt Injection (2025)
The real difficulty of production LLM API calls, why RAG is not simple lookup, agent patterns (ReAct, Plan-Execute, ReWOO), when to and when NOT to fine-tune (LoRA, QLoRA, DPO), a vector DB decision matrix, why LLM evalu
2026-04-15 · 7 min read #ai-engineering#llm#rag#ai-agent#loraANN Algorithms Complete Guide 2025: HNSW, IVF, Product Quantization, LSH — How Vector DBs Really Work
How Pinecone, Weaviate, Qdrant, and pgvector find similar items in milliseconds across hundreds of millions of vectors. A deep dive into the math and implementation of HNSW, IVF, Product Quantization, and LSH.
2026-04-15 · 14 min read #ann#hnsw#ivf#product-quantization#lshAdvanced Prompt Engineering Complete Guide 2025: CoT, ToT, Self-Consistency, Meta-Prompting
Advanced Prompt Engineering! Chain-of-Thought(CoT), Tree-of-Thought(ToT), Self-Consistency, ReAct, Meta-prompting, Structured Output(JSON Mode), System Prompt design, Few-shot optimization, Prompt Chaining, Evaluation(au
2026-04-14 · 21 min read #prompt-engineering#chain-of-thought#tree-of-thought#few-shot#system-promptAdvanced RAG Pipeline Complete Guide 2025: Chunking Strategies, Re-ranking, Agentic RAG, Evaluation
Everything about Advanced RAG! Document chunking strategies (Semantic/Recursive/Agentic), re-ranking (Cohere/ColBERT/Cross-encoder), Query Transformation, Agentic RAG (Self-RAG/CRAG), Multi-modal RAG, Evaluation (RAGAS/T
2026-04-13 · 23 min read #rag#chunking#reranking#agentic-rag#evaluationLiteLLM Complete Guide 2025: Unify 100+ LLMs with a Single API Proxy Server
Everything about LiteLLM! 100+ LLM unified API, OpenAI-compatible proxy server, cost tracking/budget management, load balancing/fallback, model routing, virtual keys, rate limiting, Guardrails, production deployment (Doc
2026-03-25 · 18 min read #litellm#llm#api#proxy#openaiBFCL Benchmark Complete Guide 2025: Tool Calling Evaluation, Leaderboard Analysis, Model Comparison
Everything about BFCL (Berkeley Function Calling Leaderboard)! Benchmark categories (Simple/Multiple/Parallel/Relevance/AST), evaluation metrics, model performance comparison (Claude/GPT/Gemini/Llama), running your own e
2026-03-25 · 22 min read #bfcl#benchmark#tool-calling#function-calling#evaluationComplete Guide to LLM Fine-tuning with Unsloth 2025: QLoRA, 4-bit Quantization, 2x Faster Training
Everything about LLM fine-tuning with Unsloth! QLoRA/LoRA theory, 4-bit quantization (bitsandbytes), Unsloth 2x speed secret, Llama 3/Mistral/Qwen fine-tuning, data preparation, training config, VRAM optimization, GGUF/G
2026-03-25 · 17 min read #unsloth#llm#fine-tuning#qlora#loraComplete Guide to Korean LLM Training Data: Hugging Face Datasets, Preprocessing, and Quality Control
Everything about LLM training data! Hugging Face datasets (types/loading/conversion), Korean data collection (crawling/synthetic/translation), preprocessing (tokenization/cleaning/dedup), Instruction Tuning formats (Alpa
2026-03-25 · 23 min read #llm#training-data#huggingface#dataset#korean-nlpAI Agent Development Complete Guide 2025: Tool Calling, ReAct, Multi-Agent, and MCP
Everything about AI Agent development! Tool Calling (Function Calling) principles, ReAct pattern, Multi-Agent architectures (CrewAI/AutoGen/LangGraph), MCP (Model Context Protocol), Agent framework comparison, Tool Calli
2026-03-25 · 26 min read #ai-agent#tool-calling#function-calling#react-pattern#multi-agentMCP Complete Guide: Why Model Context Protocol Became the USB-C of AI with 97M Downloads
Built by Anthropic, adopted by OpenAI and Google, donated to Linux Foundation. MCP hit 97M monthly downloads — the complete guide to building MCP servers, understanding the architecture, and joining the AI agent revoluti
2026-03-22 · 21 min read #mcp#ai#anthropic#openai#googleAI Prompt Engineering Complete Guide: 30 Techniques Every Developer Should Know
From Zero-shot to Chain-of-Thought, Tree-of-Thought, ReAct, Few-shot, and Self-Consistency — 30 prompt engineering techniques for developers with practical examples. How to get maximum performance from Claude, GPT-4, and
2026-03-22 · 25 min read #prompt-engineering#ai#llm#chatgpt#claude