Description

You will design and build agentic AI systems that integrate reasoning, retrieval, and secure action execution.

Responsibilities

• Design and implement tool-calling agents using MCP protocol for structured reasoning and secure action execution.
• Build evaluation frameworks and self-correction loops for production LLMs, managing prompt synthesis and response validation.
• Connect agents to observability and incident management systems to enable automated diagnostics and remediation.
• Define objective functions linked to reliability and risk reduction by translating operational challenges into agentic roadmaps.
• Build validator models and enforce deterministic fallbacks, circuit breakers, and rollback strategies.

Required Skills

• 5+ years of software development experience in Python (preferred), C/C++, Go, or Java.
• 3+ years designing and launching production ML systems, covering deployment, evaluation, and monitoring.
• Practical experience with LLMs: API integration, prompt engineering, fine-tuning, and building RAG applications with tool-using agents.
• Familiarity with commercial and open-source LLMs (OpenAI, Gemini, Llama, Qwen, Claude).
• Strong foundation in applied statistics, machine learning concepts, algorithms, and data structures.
• Experience with Python, Java, C/C++, Go.
• Familiarity with AWS services like Lambda, S3, DynamoDB, Redshift.
• Experience with infrastructure-as-code tools such as CloudFormation or Terraform.

Preferred Skills

• Experience with ECS/EKS and SageMaker.
• Experience with platform modernization and performance engineering initiatives.

Key Skills

Education

Any Gradute