Moh – AI agent orchestration, Python, LLM

An AI-powered incident intelligence and enterprise automation framework integrated with ITSM workflows to reduce operational noise, automate escalation processes, and accelerate incident response in production environments. The platform combined GenAI capabilities with incident management, RCA support, CMDB governance, and workflow automation to improve operational efficiency, compliance, and decision-making across enterprise support operations

• Designed and implemented an AI-assisted incident response and governance framework integrating Freshservice ITSM, enabling intelligent aggregation of recurring incidents, automated escalation to Major Incident (MI) and Problem tickets, and enforced linkage to Emergency / Pre-approved Change (EC/PAC) workflows for compliant remediation;
• Implemented an AI-driven RCA and decision-support layer to analyze incident records, generate contextual root cause insights, and recommend remediation actions in a human-in-the-loop model, accelerating triage and response time;
• Enabled intelligent incident correlation and pattern detection using historical data and recurrence thresholds, reducing operational noise and prioritizing high-impact issues;
• Built end-to-end AI platforms spanning ingestion, embedding, vector retrieval, orchestration, and inference, supporting scalable LLM-driven applications;
• Implemented LLM evaluation frameworks using BLEU, ROUGE, semantic similarity, and retrieval metrics (precision/recall), improving response quality, grounding, and reducing hallucinations;
• Designed CMDB lifecycle automation and data governance patterns, improving asset consistency, auditability, and downstream automation reliability;
• Delivered production AI systems including:

1. Job-matching and talent intelligence platform using LangChain, Pinecone, and microservices;
2. AI learning assistant processing TB-scale datasets using S3, EMR, SageMaker, and Bedrock.

An enterprise AI enablement initiative focused on driving secure GenAI adoption across AWS in a highly regulated environment. The program established governance frameworks, scalable architecture patterns, and production-ready AI platform capabilities aligned with Security & Architecture Assessments (ISRA & ATG), improving operational control, developer enablement, and enterprise AI readiness.

• Implemented secure AWS Bedrock Agent and AgentCore patterns, including tool registration, action orchestration, secure API integration, agent execution boundaries, and controlled enterprise workflow automation;
• Built and refined a custom enterprise AI agent pattern, enabling secure tool calling and workflow automation across Atlassian platforms and AWS-native APIs to improve decision support, traceability, and engineering productivity;
• Built and refined enterprise AI patterns using a custom AI agent (Nova), enabling secure tool calling and workflow automation across engineering systems, integrating Atlassian tools (Confluence, Jira, Bitbucket) and AWS-native APIs to improve decision-making and productivity;
• Led Amazon Kendra retirement analysis and decommission strategy by identifying active indexes, assessing SCP and guardrail impacts, mapping access permissions, and aligning replacement patterns using lower-cost deterministic retrieval approaches, resulting in over $600K in cost savings;
• Defined enterprise AI governance and onboarding patterns across AWS services and S3 Tables, aligning service enablement, ownership models, and approval processes with ISRA and ATG governance frameworks;
• Led enterprise AI enablement on AWS across service onboarding, GenAI architecture, platform governance, and production readiness in a regulated insurance environment;
• Contributed to developer platform and service catalog enablement, including reusable product patterns, onboarding flows, and governance models for cross-account adoption across the AWS organization;
• Designed and implemented a Backstage-based internal developer portal, leveraging Catalog and plugin architecture to centralize service ownership, resource metadata, and operational context, enabling self-service discovery and improving audit, governance, and engineering workflows.

An enterprise AI and cloud platform initiative focused on enabling secure GenAI adoption and large-scale AWS automation within a regulated banking environment. The program combined AI platform engineering, cloud security, governance, and infrastructure automation to support compliant AI innovation, operational scalability, and enterprise cloud modernization.

• Pioneered GenAI enablement for capital markets by building secure RAG and LLM inference platforms using AWS Bedrock, SageMaker, and LangChain;
• Designed end-to-end AI security architecture, including data validation, PII protection, prompt controls, guardrails, and audit logging across the AI lifecycle;
• Built and refined enterprise AI patterns using a custom AI agent (Nova), enabling secure tool calling and workflow automation across engineering systems, integrating Atlassian tools (Confluence, Jira, Bitbucket) and AWS-native APIs to improve decision-making and productivity;
• Established governance patterns enabling AI use cases to pass security, risk, and compliance reviews in a highly regulated environment;
• Architected centralized VPC endpoint patterns with automated Lambda-based VPC association, enabling secure and scalable onboarding of new VPCs with full cross-region connectivity;
• Implemented AWS Security Hub auto-remediation (SHARR/ASR), enabling real-time enforcement of security controls and reducing manual intervention across environments;
• Built lifecycle automation for cleanup of unused and orphaned CloudWatch alarms and log groups, reducing operational overhead and cloud costs;
• Delivered cloud automation for resource ownership discovery while enforcing enterprise-wide tagging strategy with automated validation and remediation, improving governance, cost allocation, and resource traceability;
• Worked across platform, partner, and security stakeholders to define enterprise operating models, reusable guardrails, and production intake requirements for new AWS services.

A cloud security and governance initiative supporting financial trading platforms within a Tier 1 investment bank. The engagement focused on security architecture oversight, risk governance, compliance alignment, and secure cloud adoption across AWS-based trading, data, and CI/CD environments.

• Led security architecture (SecArch) reviews across AWS-based platforms, ensuring secure design for trading systems, data pipelines, and CI/CD environments;
• Designed and enforced secure API and service-to-service communication patterns using OAuth2, OIDC, IAM, and KMS encryption;
• Established audit, logging, and monitoring frameworks aligned with NIST, ISO 27001, and PCI standards for regulated financial workloads;
• Conducted security onboarding and risk assessments of AWS services, including Macie, Inspector, Athena, and Qualys, ensuring compliance with enterprise security controls;
• Defined strategies for protecting sensitive financial and PII data, including access governance, encryption, and secure API exposure models;
• Acted as design authority for cloud security, influencing platform onboarding decisions and production readiness standards.