AI Solutions Engineer

In this role, you will: - Design and develop agentic AI systems by creating tool-calling agents that incorporate retrieval, structured reasoning, and secure action execution while adhering to MCP protocol. Establish strong guardrails for safety, compliance, and minimal access permissions. - Productionize large language models by creating evaluation frameworks for open-source and foundational LLMs; develop retrieval pipelines, prompt synthesis, response validation, and self-correction loops uniquely suited for production settings. - Integrate agents within runtime ecosystems by connecting them to observability, incident management, and deployment systems to facilitate automated diagnostics, runbook execution, remediation, and incident summarization with comprehensive traceability. - Collaborate directly with production engineers and application teams to transform production challenges into agentic AI roadmaps, defining objective functions connected to reliability, risk mitigation, and cost efficiency while ensuring auditable and business-aligned results. - Implement safety, reliability, and governance measures by developing validator models, adversarial prompts, and policy checks; enforce deterministic fallbacks, circuit breakers, and rollback strategies; and carry out ongoing evaluations for usefulness, accuracy, and risk management. - Enhance scalability and performance by optimizing costs and latency through prompt engineering, context management, caching, model routing, and model distillation while utilizing batching, streaming, and parallel tool calls to meet stringent service level objectives under real-world conditions. - Create a Retrieval-Augmented Generation (RAG) pipeline by curating domain knowledge, building a data quality validation framework, and establishing feedback and milestone frameworks to maintain knowledge relevance. - Set high standards by driving design reviews, enforcing rigorous experimentation, and promoting high-quality engineering practices; mentor peers on agent architectures, evaluation methodologies, and safe deployment protocols.