Jul 21, 2026

Oracles, Agents, Duration, and Inference Ops

A daily crypto, finance, and AI vocabulary puzzle covering data feeds, autonomous workflows, rate sensitivity, and model serving.

Oracle Data

These words describe how blockchain systems pull, validate, and time external data.

OraclePrice FeedTWAPAttestation
  • Oracle: An oracle is a service that supplies outside information to a blockchain application.
  • Price Feed: A price feed is a data stream that publishes asset prices for apps and smart contracts.
  • TWAP: TWAP means time-weighted average price, a method that smooths out short-term spikes.
  • Attestation: An attestation is a signed statement that a piece of data or state is valid.

Agentic Workflows

These words describe how an assistant decomposes tasks, uses tools, and maintains context.

AgentTool UsePlannerContext Window
  • Agent: An agent is software that can plan, choose actions, and work toward a goal with some autonomy.
  • Tool Use: Tool use is the practice of calling external APIs or functions during task execution.
  • Planner: A planner is the part of an agent that breaks a task into steps or selects actions.
  • Context Window: The context window is the amount of text a model can process at one time.

Rate Sensitivity

These words describe how fixed-income assets respond to changes in rates and policy expectations.

Treasury BillYield CurveDurationBasis Point
  • Treasury Bill: A treasury bill is a short-term U.S. government debt instrument often used as a cash-like benchmark.
  • Yield Curve: A yield curve shows how interest rates vary across different maturities.
  • Duration: Duration measures how sensitive a fixed-income asset is to changes in interest rates.
  • Basis Point: A basis point is one-hundredth of a percentage point.

Model Serving

These words describe how inference systems trade speed, cost, and precision.

  • Inference: Inference is the process of using a trained model to produce an output.
  • Batching: Batching combines multiple requests so a system can process them more efficiently.
  • Quantization: Quantization reduces numeric precision to make models smaller or faster.
  • Throughput: Throughput is the amount of work a system can process per unit of time.