Blog
Writing
Life, tech, and everything in-between.
Handoffs and the Swarm Pattern: Peer-to-Peer Agent Transfer
A supervisor is not the only way to coordinate agents. In the swarm pattern, agents transfer control to each other through handoff tool calls. No central manager; the conversation simply moves to whichever specialist it needs to be with.
Hierarchical Teams: Supervisors of Supervisors in Multi-Agent Systems
A supervisor with five specialists works. A supervisor with twenty does not. Hierarchical teams stack supervisors, each responsible for a department of specialists, and keep a dual ledger to track plan and progress separately.
Supervisor and Router: A Central Agent That Delegates to Specialists
When a single agent carries too many tools and too many responsibilities, the fix is not a bigger prompt but a supervisor whose tools are other agents. Each specialist has its own prompt, its own tools, and its own scratchpad.
LATS: Monte Carlo Tree Search for LLM Agent Decisions
LATS brings Monte Carlo Tree Search, the algorithm behind AlphaGo, to language agents. Reasoning, acting, and planning become a single tree that the agent navigates by balancing exploration against exploitation. Accuracy climbs, cost climbs with it.
Reflexion: Verbal Self-Critique After Failure for LLM Agents
When an agent fails a task, a separate critique call can write down what went wrong in words, store the critique as a memory, and let the next attempt condition on it. No gradient updates. A model that learns from text about its own past.
Graph-of-Thought: Non-Linear Reasoning with Merge and Refine
Trees branch and backtrack but never merge or loop. Graph-of-Thought extends the search to a directed graph, adding aggregation of parallel branches and refinement of earlier thoughts. The payoff is on problems that need synthesis rather than pure search.
Tree-of-Thought: Branching Reasoning with Search for LLMs
Chain-of-thought commits to one path. Tree-of-thought explores several, evaluates them, and backtracks. On problems that need exploration rather than sequential derivation, the difference is dramatic.
Chain-of-Thought Prompting for LLM Reasoning
Asking a large language model to reason step by step before answering reliably improves performance on multi-step problems. The technique is small in the prompt and large in effect, but only on models above a certain scale.
The Agent-Computer Interface: Designing Tools for LLM Agents
Anthropic reports spending more time tuning tool descriptions than tuning prompts for a SWE-bench agent. Tool design is where agentic systems actually succeed or fail, and it pays more compounding returns than almost any other investment.
Plan-and-Execute: Two-Phase Agents That Plan First, Then Act
ReAct interleaves planning with acting at every step. Plan-and-execute produces a complete plan first, then executes it. The separation trades adaptivity for token efficiency and is often the better choice when tasks have clear structure.
The Tool-Calling Agent Loop: ReAct as It Actually Ships
The 2022 ReAct paper parsed thought-action-observation from free text. Production agents today use native tool-calling, a step budget, input and output guardrails, and explicit handoff handling. The simple loop earns its hardening.
ReAct: Reasoning and Acting in One LLM Loop
The foundational agent pattern from Yao et al. interleaves a short reasoning trace with a single tool call, then folds the observation back into the next step. Simple in shape, strong on grounded reasoning benchmarks.