16 articles
Browse guides
MCP standardises how models discover and call external capabilities, so a tool built once works across any compatible client instead of being rewritten per vendor. Introduced by Anthropic in late 2024 and adopted more broadly since, it is the closest thing the agent ecosystem has to a common connector. In practice: One MCP server for your ticketing system, usable from any MCP-aware assistant.
Multi-agent setups split a job across agents with narrow roles — research, draft, critique — on the theory that focused context beats one agent juggling everything. They genuinely help on tasks that decompose cleanly. They also multiply cost, latency, and failure modes, so reach for one only after a single agent has demonstrably failed. In practice: One agent gathers sources, a second writes, a third fact-checks against the sources.
Orchestration is the routing and control logic around the models: retries, fallbacks, cheap-model-first escalation, parallel steps, state. In production this layer, not the model, is usually what determines whether the system is reliable and affordable. In practice: Route easy tickets to a small model, escalate ambiguous ones to a reasoning model.
ReAct — reason plus act — has the model state its thinking, take one action, read the observation, and think again. Alternating the two beats reasoning everything upfront because real environments return surprises. Most agent frameworks implement some version of it. In practice: Thought: I need the invoice date. Action: search_invoices. Observation: three results. Thought: pick the most recent.
Tool use is what lets a model do things it fundamentally cannot do alone: look up current data, run exact calculations, write to a system. The model chooses the tool and the arguments; your code executes it and returns the result. Tools are also the point where a model’s mistakes become real-world actions, so permissions matter. In practice: Rather than guessing today’s rate, the model calls a currency API.
Workflow automation is the deterministic cousin of agents: you specify the steps and the model handles the parts that need judgement. It covers the large majority of business use cases and is more reliable and cheaper than an agent deciding for itself. Use an agent when the path genuinely varies; otherwise use a workflow. In practice: Email arrives → extract fields → classify → route → draft reply → hold for approval. ...