Tags

Fundamentals

llm (34)Large language models in general
transformers (6)Transformer architecture and attention
tokens (5)Tokenization, context windows, token economics
embeddings (4)Vector representations and similarity search
training (4)Pre-training, post-training, RLHF and related methods
fine-tuning (5)Adapting models to tasks (LoRA, SFT, distillation)
reasoning (17)Extended thinking, chain-of-thought, test-time compute
multimodal (16)Vision, audio, video input and output
robotics (3)Embodied AI, robot control, and robotics foundation models

Techniques

prompting (6)Prompt design and context engineering
rag (3)Retrieval-augmented generation
tool-use (5)Function calling and tool integration
structured-output (2)JSON, schemas, constrained decoding
evals (4)Evaluation methods and benchmarks
caching (8)Prompt caching and response caching
memory (3)Persistent memory for agents and assistants

Agents

agents (24)Agent design, loops, and autonomy
multi-agent (4)Orchestration of several agents
coding-agents (7)Agents that write and modify code
computer-use (3)Agents that operate GUIs and browsers

Tooling

mcp (9)Model Context Protocol servers and clients
skills (2)Packaged agent skills and instructions
sdk (6)Official SDKs and client libraries
cli (5)Command-line AI tools
ide (1)Editor and IDE integrations

Frameworks

frameworks (9)Agent and LLM application frameworks
open-source (17)Open-weight models and open-source projects

Providers

anthropic (22)Anthropic and Claude
openai (9)OpenAI and GPT models
google (17)Google and Gemini
meta (3)Meta and Llama
mistral (3)Mistral AI
other-provider (7)Any other model provider

Practices

deployment (8)Serving, hosting, latency, scaling
cost (17)Pricing and cost optimisation
safety (10)Safety, alignment, guardrails
security (15)Prompt injection, data exposure, supply chain
research (11)Research papers and results
industry (7)Business, policy, and ecosystem news