AI/Data Engineer Agentforce

Role overview A contributor-level opportunity to build production AI agents on a Salesforce-based agent platform during an early-adoption phase, where the implementation patterns are still being defined rather than inherited. You will be embedded as part of a North American client squad, working remotely from a Latin American delivery center with substantial time-zone overlap into the client's core business hours. The work centers on agent design, evaluation, and the surrounding data plumbing, not just prompt writing.

Responsibilities - Design and assemble agents inside Agent Builder, defining topics, instructions, and the boundaries between them. - Implement agent actions across Flow, Apex, prompt templates, and external APIs. - Ground agents in Data 360 and retrieval sources so responses come from data rather than from model generation. - Design guardrails against prompt injection, data leakage, and out-of-scope action execution. - Build evaluation sets and regression tests for agent behavior and run them before each change. - Instrument agents for observability, capturing what was asked, retrieved, executed, and failed. - Prepare structured and unstructured data for retrieval, including chunking and metadata strategy. - Collaborate with the Data 360 team on freshness, identity, and consent constraints. - Analyze live transcripts and deflection data to drive evidence-based improvements. - Advise stakeholders clearly when a use case is not a good fit for an agent.

Requirements - Bachelor's degree in Computer Science, Data Science, Software Engineering, or a related field, or equivalent experience. - Four or more years in software or data engineering, with hands-on production experience building LLM-backed applications. - Practical prompt engineering experience, including systematic prompt testing. - Working understanding of AI risk areas such as hallucination, injection, PII exposure, and action authorization. - Familiarity with Agentforce components including Agent Builder, topics and actions, Prompt Builder, and the Einstein Trust Layer. - Enough Salesforce development depth in Flow and Apex to build and debug agent actions. - Comfort with retrieval-augmented generation concepts such as embeddings, chunking, vector retrieval, and grounding evaluation. - Python skills for data preparation, evaluation harnesses, and API work. - Experience using Salesforce Data 360 as a grounding source. - Professional English sufficient to run requirements and design conversations directly with North American stakeholders. - Permanent authorization to work in your country of residence, with no visa sponsorship available.

Nice to have - Salesforce Certified AI Specialist or Agentforce Specialist credentials.

Benefits and work setup - Full-time contractor agreement, fully remote. - Delivered from a Latin American delivery center into a North American client team. - Employment relationship held by the delivery organization; you sit inside the client's squad under delivery leadership. - Equal opportunity employer committed to diversity and inclusion.