By Chandan AgarwalYou will learn how to structure, validate, and deploy Python scripts as managed Code Extensions within Data 360. The guide also covers connecting these scripts to batch data transforms and monitoring execution through the platform.
Northstar Outfitters faced inconsistent product data across commerce systems and needed a trusted catalog. Native visual transforms could not handle the required standardization, enrichment, and scoring logic, prompting the use of Data 360 Code Extension to run modular Python and PySpark scripts directly on Salesforce infrastructure.
The solution relies on isolated runtime environments where scripts read from and write to designated Data Lake Objects using explicit permissions defined in a configuration file. Developers validate logic locally against sampled data before packaging the script, dependencies, and tests into a deployable archive.
The guide details both Setup interface and Salesforce CLI workflows for uploading packages, selecting compute sizes, and linking the extension to a batch data transform. Execution, scheduling, and logging are handled by the platform, while operators can trigger runs and review history through REST API calls.
By Raveendrnathan Loganathan and Tobias MuehlbauerReaders will understand how the Agent Context Engine retrieves, reconciles, and packages authorized evidence into token-efficient Context Packs for AI agents. The piece also explains how Data 360 separates runtime state from durable memory to maintain governance across models and channels.
Enterprises face a significant context problem when feeding AI agents because loading all enterprise data into prompts wastes tokens, increases latency, and risks exposing sensitive information. Salesforce addresses this with Data 360, which acts as a shared runtime foundation to compile current, relevant, and authorized evidence without duplicating the entire data estate across platforms.
The Agent Context Engine operates bidirectionally to assemble token-fit Context Packs for outbound agent turns while returning typed interaction traces for auditing on the inbound path. It manages four distinct memory states, separating temporary working context and short-term traces from approved long-term memory and distilled learnings to ensure only validated facts persist across sessions.
The platform supports a distributed enterprise model where data remains authoritative in external systems like Databricks and Snowflake while participating through ingestion or Zero Copy federation. Pro-code agents can integrate via SDKs and API adapters, while third-party assistants connect through standard protocols, all unified under the Enterprise AI Harness for consistent governance, security, and observability.