IAM Architect & AI/ML Engineer specializing in LLM applications, RAG pipelines, agentic AI, and enterprise identity — OAuth2/OIDC/SAML, MFA, device trust, and ML in authentication workflows.
I design and ship production-grade identity platforms and ML systems, not prototypes.
Brain warm-up for ML students • Linear Algebra Fundamentals
Master vector operations step-by-step with interactive visuals
Choose a demo mode to start your vector journey
I'm an IAM architect and quality engineering leader with 9+ years building enterprise identity platforms — OAuth2/OIDC/SAML, MFA, directory integration, device trust, and risk-adaptive authentication — at Acceptto and SecureAuth, where automation-first quality practices cut release defect density up to 85%. Today I'm architecting a confidential, pre-launch integrated identity platform with agentic AI assurance, as sole quality owner building toward a Quality Engineering Center of Excellence.
AI/ML is a genuine passion alongside this work — I build production-ready AI systems, including LLMs, RAG, tool-calling agents, and classical ML, held to the same bar: fail-closed step-up MFA, eval-gated releases, and validation through CI. I focus on problems where IAM architecture, quality engineering, and AI/ML intersect.
OAuth2/OIDC/SAML, MFA, directory sync, device trust, RBAC, risk-adaptive auth
Classical ML, deep learning, embeddings, ML in authentication workflows
Multi-phase pipelines, RAG, grounding, tool-calling
Fail-closed gates, step-up auth, structured reasoning, safety layers
Go, FastAPI, microservices, IdP protocols, secure APIs
CI/CD, eval harnesses, test automation, model lifecycle, observability
Go-based identity and access management platform for enterprise authentication flows. Provides OAuth2/OIDC/SAML protocol surfaces, MFA orchestration, device trust and fingerprinting, session and token lifecycle, and hardened credential handling — designed as a modular IdP foundation for workforce and client identity.
Production-oriented agentic AI service for identity, assurance, and admin workflows. Combines domain-specialist agents, RBAC-scoped tools, retrieval-augmented generation, and fail-closed safety gates — including step-up authentication, citation enforcement, and versioned prompts — to augment identity platforms through HTTP adapters.
Cross-platform Flutter mobile authenticator for enterprise MFA and device-bound identity. Supports EC P-256 device key generation, push and inline MFA approval flows, offline TOTP, secure pairing with backend identity services, and release hardening including TLS pinning — extending workforce authentication to mobile endpoints.
Real-time AI co-pilot desktop app for high-stakes live conversations — client consulting calls, discovery meetings, negotiations, and other professional dialogue where you need private, in-the-moment guidance. Captures remote participant audio from the call, transcribes locally with Whisper and RNNoise, and fires parallel directional, depth, and clarifying LLM threads in a stealth overlay invisible to screen-share. RAG over session context via sqlite-vec keeps guidance grounded in your brief, notes, and domain material. Phase 1 ships the job-interview domain (mock sessions, rehearsal, preferred answers) as the first vertical.
AI-powered job-search platform. Build a persistent master resume by voice (Story Mode + coached interview) or upload, tailor it to any job description through a four-phase agent pipeline, generate cover letters, check job-fit scores, search for matching listings, and track every application — all in one place.
End-to-end fraud detection pipeline for credit card transactions. Covers EDA with timezone-aware feature engineering, scalable PySpark preprocessing, supervised ML, deep learning (MLP/ResNet/LSTM), transformer-based models (FT-Transformer), hybrid ensemble stacking, and a real-time Streamlit inference app with a two-stage risk-tiering system (Auto-Block / Review / Cleared).
A comprehensive ML solution predicting customer term deposit subscriptions using 8+ algorithms with hyperparameter tuning, achieving 60% F1-Score improvement over baseline models.
Production-ready machine learning solution for predicting customer churn with 84.77% ROC-AUC. Features 16 optimized models, SHAP interpretability, Optuna hyperparameter tuning, and comprehensive business insights.
Production-ready Flask web application providing comprehensive REST API and interactive UI for the Movie Agent Service. Features intelligent movie discovery through natural language queries, real-time chat interface with conversation history, movie poster image analysis using computer vision, interactive movie quizzes, and formatted statistics display. Implements encrypted API key storage with local configuration management, session-based memory isolation for multi-user support, and automatic query classification for seamless routing.
Production-grade core library for AI-powered movie discovery and analysis. Implements RAG-based semantic search using FAISS vector store, tool-calling agent architecture with LangChain (chosen over ReAct for reliability), computer vision poster analysis with BLIP model, interactive quiz generation, and session-based memory management. Built with Python following SOLID principles, dependency injection, and factory patterns for maintainability and extensibility.
Advanced regression pipeline predicting satellite signal quality from weather conditions using multiple ML algorithms and real-time API data.
Machine learning project predicting passenger survival using ensemble methods and feature engineering techniques.
Learn the fundamentals of Natural Language Processing with interactive examples and hands-on exercises.
Master Naive Bayes classification with interactive weather prediction demo and comprehensive mathematical explanations.
Complete hands-on course with Python implementations and real-world examples covering introduction, regression, and classification.
Asset managers are not just fighting yesterday's threats. AI has handed attackers capabilities that scale infinitely. Quantum computing is on the horizon. This is the story of what is actually changing, and what the financial industry needs to do before the window closes.
Ramp data shows U.S. firms paying DeepSeek directly for cheaper inference — but hosted API use routes data to China under PRC law. A technical breakdown of cost pressure, documented risks, self-hosting nuance, and what security claims hold up.
Let's discuss IAM architecture, AI/ML systems, and production identity platforms.
Get In Touch