Lead Agentic AI Engineer
C
CodeRoadAI software
Location: Latin AmericaFull-TimeLead
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Languages
- Fluent written and spoken professional English
- Experience
- 7+ years of hands-on enterprise software engineering experience; 3+ years of direct production experience building, evaluating, and scaling LLM-powered applications and autonomous multi-agent workflows.
- Required Skills
- PythonJavaSalesforceLangChain
Requirements
- Have 7+ years of hands-on enterprise software engineering experience using Python, TypeScript/Node.js, Java, or Apex.
- Have 3+ years of direct production experience building, evaluating, and scaling LLM-powered applications and autonomous multi-agent workflows.
- Demonstrate expert proficiency with LangGraph or LangChain, Anthropic/Claude APIs, and MCP.
- Have experience with observability tools such as LangSmith, Langfuse, or OpenTelemetry.
- Have hands-on experience with automated LLM evaluation suites such as Promptfoo, Ragas, or LLM-as-a-judge patterns as automated test gates.
- Bring diagnostic debugging skills to isolate non-deterministic LLM behavior, context degradation, and latency bottlenecks.
- Be able to translate complex AI mechanics into architecture roadmaps for technical and non-technical stakeholders.
- Have fluent written and spoken professional English for direct client consultation.
- Salesforce API integration experience, including REST, Composite, or Bulk APIs, SOQL/SOSL, Platform Events, and Connected Apps, is a nice-to-have.
- Familiarity with Salesforce Data Cloud, vector search, or Agentforce patterns is a nice-to-have.
Responsibilities
- Architect stateful multi-agent coordination graphs using LangGraph or custom DAG loops, including task decomposition, supervisor orchestration, human-in-the-loop escalation, and error recovery.
- Build standardized MCP servers and connectors to securely expose Salesforce tools, objects, custom actions, and operational context.
- Optimize query routes and context engineering across model tiers to improve latency, accuracy, and token spend.
- Design automated evaluation harnesses with Promptfoo, Ragas, or DeepEval and integrate them into CI/CD pipelines.
- Implement observability and defense-in-depth security using LangSmith or Langfuse, input sanitization, field-level security, and PII scrubbing.
- Establish API interface contracts, specification-driven development patterns, and architectural blueprints for engineering teams and executive stakeholders.
View Full Description & ApplyYou'll be redirected to the employer's site
