InTowards AIbyErez Azaria·Jun 22Emergent Behavior in Agentic SwarmsWhat building an agentic swarm taught me about control, delegation, and letting go
InTowards AIbyErez Azaria·Mar 20Cutting Claude Code Costs Without Sacrificing QualityHow to use Claude code with mixed native and Ollama subagent modelsA response icon1A response icon1
InTowards AIbyErez Azaria·Jan 26From Spatial Navigation to Spectral Filtering: A New Transformer Inference FrameworkRethinking inference via signal processing explains why cosine similarity dominates, norms grow, and how next tokens are selected.
InTowards AIbyErez Azaria·Dec 15, 2025The Atomic Traits of LLMsQuantifying the Behavioral Signatures of Foundation ModelsA response icon1A response icon1
InTowards AIbyErez Azaria·Jun 4, 2025The Anchor That Almost WasSimulated Introspection in Large Language ModelsA response icon3A response icon3