A daily crypto, finance, and AI vocabulary puzzle covering reserves, execution costs, production inference, and agent safety.
Stablecoin Balance Sheets
These terms describe reserve support, verification, cash-out mechanics, and the target price relationship.
- Reserve Backing: Reserve backing refers to the assets held to support a stablecoin's target value.
- Attestation: An attestation is a report or statement that helps verify reserves or other claims.
- Redemption: Redemption is the process of swapping a token back for the asset or value it tracks.
- Peg: A peg is a target price relationship that a stable asset tries to maintain.
Order Execution
These terms describe how orders are placed, how markets absorb size, and why fills can move price.
- Limit Order: A limit order only executes at the specified price or a better one.
- Market Order: A market order executes immediately against the best available liquidity.
- Bid-Ask Spread: The bid-ask spread is the difference between the highest buy price and the lowest sell price.
- Slippage: Slippage is the difference between the expected trade price and the actual fill price.
Model Serving
These terms describe runtime work that makes a model fast, efficient, and predictable in production.
- Inference: Inference is the process of using a trained model to generate outputs for new inputs.
- Batching: Batching combines multiple requests so they can be processed more efficiently.
- Quantization: Quantization reduces numerical precision to make models smaller and faster to run.
- Latency: Latency is the delay between a request and the start or completion of a response.
Agent Controls
These terms cover planning, sandbox boundaries, memory, and offline testing for agent workflows.
- Tool Calling: Tool calling is the act of invoking a function, API, or service from an AI workflow.
- Sandbox: A sandbox is a controlled environment that limits what code or an agent can affect.
- Memory: Memory is stored context that can help an agent reuse facts or preferences later.
- Evals: Evals are repeatable tests used to measure whether a model or agent behaves as expected.