Note: The job is a remote job and is open to candidates in USA. Kani Solutions Inc is seeking a skilled Agentic AI Platform Engineer to help build and scale a centralized agentic AI ecosystem. The role involves developing secure, scalable AI orchestration services and enabling reusable AI capabilities for engineering teams.

Responsibilities

• Develop and maintain orchestration services using AWS Lambda, SQS, and event-driven architectures to manage AI agent workflows
• Build integrations with GitLab CI/CD pipelines, enabling AI-powered automation within build and deployment stages
• Create reusable serverless tools and shared service components that can be leveraged by multiple engineering teams
• Design and implement AI knowledge repositories using AWS Bedrock Knowledge Bases, OpenSearch, and S3 to improve contextual reasoning for AI agents
• Implement governance and security controls including guardrails, sensitive-data filtering, access restrictions, and prompt validation mechanisms
• Optimize AI usage through model routing, semantic caching, token management, and cost-aware processing strategies
• Establish observability and monitoring standards using CloudWatch, centralized logging, tracing, and audit reporting for AI interactions and platform usage
• Collaborate with platform and engineering teams to improve extensibility, scalability, and developer experience

Skills

• 3–7 years of professional software engineering experience
• Strong proficiency in Python and AWS SDKs for building production-grade serverless applications
• Hands-on experience with AWS services such as Lambda, API Gateway, SQS, EventBridge, IAM, Secrets Manager, and CloudWatch
• Experience working with Amazon Bedrock, including Agents, Action Groups, Knowledge Bases, and Guardrails
• Practical experience integrating Large Language Model APIs such as Bedrock, OpenAI, Anthropic, or similar platforms
• Good understanding of CI/CD concepts and hands-on exposure to GitLab CI pipelines
• Strong awareness of application security, including IAM best practices, secrets handling, prompt injection prevention, and secure AI workflow design
• Experience implementing logging, tracing, monitoring, and debugging solutions in distributed systems
• Experience with vector databases, retrieval-augmented generation (RAG), embeddings, or OpenSearch
• Familiarity with AI cost optimization strategies including semantic caching, inference routing, and token utilization tracking
• Experience building internal developer platforms, automation tooling, or reusable engineering frameworks
• Exposure to agentic AI frameworks and orchestration patterns

Company Overview