parallel-cli

Automated web research and data enrichment is executed through a structured command-line interface that supports asynchronous job management and structured JSON output. It enables complex, multi-step information gathering without manual browser interaction.

Existing CSV or JSON records can be enriched by ingesting data files to automatically append web-discovered information like revenue, leadership, or entity details. This structured enrichment process adds depth to your original dataset.

Entity discovery is handled by the FindAll workflow, which performs web-scale discovery of startups, companies, or specific entities based on complex criteria. It automates the search process to find matching organizations across the web.

An API key is required to interface with the Parallel web-native research stack and execute asynchronous research jobs. You also need the parallel-cli binary installed to run the commands and manage processor tiers.

Asynchronous job management allows deep research tasks to execute in configurable processor tiers for varying levels of depth. It supports context-aware follow-up queries, enabling complex information gathering without blocking terminal execution.

Structured JSON output is natively supported for deep research reports and data enrichment tasks. This format ensures the results are machine-readable and ready for downstream processing or integration into other data pipelines.