Senior AI Engineer - Agentic AI & Platform Assessment (LATAM)
Remote
Job Summary
Review existing LLM, AI-agent, and MCP-based capabilities across the multi-tenant digital marketplace platform, tracing workflows from prompt input through model invocation and downstream data writes. Identify prompt-injection attack surfaces, validate LLM outputs before database writes, and assess tool permissions, tenant isolation, and auditability of AI actions. Evaluate structured-output patterns, hallucination handling, and AI coding agent usage within the SDLC to recommend measurable productivity and reliability metrics. Produce evidence-backed findings with severity ratings and remediation strategies, contributing to the hardening scorecard and 30/60/90 roadmap. Collaborate with Solution Architects and Senior Full-Stack Engineers during findings validation and prioritization.
Required Qualifications
- 7+ years of software engineering experience
- strong hands-on Generative AI / LLM engineering experience
- Production experience integrating LLMs such as Claude, OpenAI models or equivalent
- Strong experience with LLM APIs, structured outputs, function/tool calling and agentic workflows
- Practical experience with Model Context Protocol (MCP) or comparable agent/tool integration patterns
- Strong understanding of prompt injection, LLM security and AI application attack surfaces
- Experience with schema validation, API integration and application/database write paths
- Strong knowledge of software engineering, REST APIs and modern cloud-native applications
- Experience reviewing existing production code and identifying technical, security and governance risks
- Understanding of multi-tenant SaaS security and authorization patterns
- Hands-on Generative AI / LLM engineering
- Agentic AI and tool-calling architecture
- Prompt-injection and AI security knowledge
- LLM output validation and governance
- Strong code-review and troubleshooting skills
- Ability to collaborate with architects and application engineers
- Ability to work during U.S. business hours
Desired Qualifications
- Experience with Claude / Anthropic SDK and Claude Code
- Experience with Supabase, PostgreSQL and Edge Functions
- Experience with RAG, intelligent matching or recommendation systems
- Experience building AI-enabled marketplace or workflow platforms
- Familiarity with AI governance for government, public-sector or regulated environments
- Exposure to MCP servers and AI-enabled API ecosystems
- Experience measuring AI-assisted engineering productivity
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.