Loading…
How AI Rebuilt Salesforce’s Decades-Old Localization Pipeline
SalesforceScott Nyberg
Summary
Salesforce needed to deliver major releases across 34 languages as product localization volume surged by more than 35% under fixed budgets and unchanged delivery windows. To address these scaling constraints, the engineering team replaced their legacy translation workflows with an LLM-powered orchestration pipeline integrating prompt engineering, context engineering, and multi-stage validation. The architecture transforms existing style guides, glossaries, and terminology databases into structured context, routing interface strings through specialized prompt stages—up to 85 stages for some languages—to enforce product context, grammar, and branding rules. Combined with an automated validation loop spanning AI editing, AI validation, and human review, this system reduced localization costs by 50 to 90% while accelerating turnaround times across enterprise releases.
Context
Salesforce releases ship in up to 34 languages with the same functionality, quality, and schedule as the English release. Product-related localization volume surged by over 35% between releases, while translation budgets and release windows remained fixed, rendering human-only translation and basic machine translation insufficient for scaling.
Approach / What changed
The engineering team built an LLM orchestration pipeline combining prompt engineering, context engineering, and multi-stage validation. Legacy style guides, glossaries, and terminology databases were converted into structured context for models. Strings move through sequenced prompts establishing grammar, branding, and product context, followed by AI editing, AI validation, and human review.
Takeaways
- Depending on the target language, a single UI translation can move through as many as 85 specialized prompt stages before reaching production.
- The LLM-powered localization pipeline reduced localization costs by 50 to 90%, depending on the specific workflow, while accelerating turnaround times.
- Legacy style guides, glossaries, and terminology databases were restructured as contextual data used by models to generate, edit, and validate translations.
Related reading
How Salesforce Built an Agentic Engineering Enablement Strategy for Thousands of Software Engineers
Salesforce addressed the challenge of scaling agentic engineering across thousands of software developers by focusing on organizational learning rather than tooling. While individual engineers initially developed disparate, uncoordinated workflows, the Technology, People, Innovation, and Learning (TPIL) team unified these approaches into the Proficiency Level (PL) Framework. The framework maps developer growth across four mindset-driven stages: AI-Assisted, AI-Validating, AI-Orchestrating, and AI-Native. To facilitate progress across these stages, TPIL introduced global AI camps, weekly working sessions, and manager coaching guides. Rather than tracking course completions or telemetry scores, Salesforce measured transformation through behavioral indicators, such as shifts toward validating agent output, designing multi-agent workflows, and asking higher-level questions regarding human judgment.
Scott NybergHow Agentforce-Powered AI Security Workflows Accelerate Incident Response
Salesforce engineers expanded Security Center from a basic conversational interface into an Agentforce-powered stateful investigations platform for incident response. Managing enterprise security incidents required handling long-running lifecycles, disparate telemetry formats, and finite LLM context windows that risk reasoning pocket collapse. To prevent overflowing context limits, the architecture partitions telemetry into contextual segments and uses AI-driven summarization pipelines before feeding data to the agent. Because non-deterministic LLM outputs break traditional deterministic testing, the team implemented AI-driven evaluation pipelines using simulated customer interactions to evaluate response behavior rather than exact wording. Ongoing work focuses on grounding public foundation models with curated, Salesforce-owned security knowledge to improve platform-specific remediation guidance.
Scott NybergWhy AI-Generated Code Is Easy but Engineering Trust Is Hard
At Salesforce, engineering teams discovered that AI coding agents could produce internally consistent code and passing test suites that still failed to meet actual requirements. To ensure correctness before writing code, the team developed an agentic Spec-Driven Development workflow centered on explicit specifications and gated verification. The system separates evidence-based repository lookups handled by agents from subjective judgment calls escalated to human engineers. Implementation plans must cite repository evidence reviewed by a Skeptic Agent, while a Compliance Matrix traces each success criterion directly to executable test proof. Finally, a multi-agent review system evaluates the implementation under an asymmetric rule allowing an independent judge to downgrade passing checks but never override failed gates.
Scott NybergHow to Evaluate Production AI Agents: Measure System Outcomes, Not Conversations
Evaluating AI agents solely on conversational transcripts fails when agents must invoke tools to modify external production systems. Language models can convincingly narrate that an action occurred while completely omitting the underlying tool call, leaving database or billing records unchanged. Outcome-based evaluation addresses this divergence by treating external system mutations rather than generated dialogue as the actual source of truth. Frameworks such as CRMAgentBench implement persistent, stateful environments with strict all-or-nothing validation, verifying tool arguments, execution sequence, final record modifications, and the total absence of unauthorized actions. Furthermore, measuring repeated success across independent executions using pass^k metrics exposes operational variance and inconsistency that standard single-shot scores and pass@k metrics conceal.
Scott Nyberg