RECRUITER-SPECIFIC PORTFOLIO VIEW

Agentic AI Engineering

Principal Agentic AI Architect

Governed agents, RAG, tool use, orchestration, evaluation, identity, and human authority boundaries.

This view focuses on multi-framework agent implementation, MCP and enterprise tool integration, delegated authorization, RAG quality, and evidence-backed agent behavior.

SELECTED PROOF

Projects most relevant to this role.

6 systems selected from the curated engineering catalog.

01 · Agentic AI Framework EngineeringLab complete

Multi-Framework Agentic Evidence Lab

The same governed RAG workflow implemented four ways to compare framework behavior fairly rather than compare unrelated demos.

A completed Python engineering lab implementing a shared evidence-review workflow across LangChain, LangGraph, Strands, and ADK-style patterns with common schemas, tools, datasets, benchmark questions, trace envelopes, scoring rubrics, structured outputs, and human-review routing.

PythonLangChainLangGraphStrandsADKPydantic
02 · Agent Identity & AuthorizationValidated

Brokered Agent Delegation & Enterprise API Security

Secure cross-application agent action without broad standing privileges or independent superuser authority.

A deterministic Python implementation of user-bound delegation, OAuth 2.0 token-exchange concepts, OIDC claims validation, policy-as-code authorization, scoped delegated tokens, API-side enforcement, abuse-case testing, and evidence records for allow and deny decisions.

PythonOAuth 2.0OIDCPolicy as CodeRegoPytest
03 · LLM Quality & Release EngineeringLive

LLM EvalOps & Release Evidence Platform

Deterministic evaluation, RAG quality, safety red-team testing, regression gates, and fail-closed release evidence.

A production-style Python evaluation platform with provider abstraction, JSONL datasets, prompt boundaries, model comparison, RAG metrics, safety checks, experiment tracking, model aliases, CI gates, and a public release dashboard.

PythonPytestRAG EvaluationGitHub ActionsOpenAI AdapterJSONL
04 · Agent Identity RuntimeArchitecture

Zero Trust Runtime for AI Agents

Build-ready Agent Identity Broker issuing JIT, least-privilege credentials with deterministic OPA enforcement.

A FastAPI and OPA runtime skeleton for machine identity, structured intent, short-lived scoped tokens, and audit evidence at the agent-to-tool boundary.

PythonFastAPIOPAKubernetesJIT CredentialsMCP
05 · Agent InteroperabilityLab complete

MCP Interoperability Foundations Lab

Systems-thinking implementation path for standardizing how agents discover and invoke enterprise tools across platforms.

A focused learning lab explaining tool contracts, adapters, cross-platform routing, and the role of MCP in reusable enterprise agent integration.

MCPTool ContractsAdaptersAWSAzureSaaS APIs
06 · Enterprise AI IntegrationArchitecture

Enterprise AI Integration Architecture Lab

Enterprise AI gateway pattern connecting identity, RAG, policy, CMDB, ticketing, audit, approval, and observability.

A modular integration architecture for regulated enterprise workflows, showing how AI can retrieve context, evaluate policy, call systems, and create auditable evidence.

AI GatewayRAGCMDBTicketingPolicy EngineAudit Service

NEXT STEP

Need a focused architecture conversation?

I can walk through the implementation choices, tradeoffs, evidence, boundaries, and production-hardening path behind these systems.

Contact Ola