[{"content":"Description AIPAF is a Python CLI and an assessment methodology for evaluating enterprise AI projects against NIST AI RMF 1.0, ISO/IEC 42001:2023, and the EU AI Act.\nIt scores a project across six dimensions — strategic alignment, risk classification, data readiness, technical feasibility, ethical \u0026amp; responsible AI, and operational readiness — through a deterministic scoring engine that produces a GREEN / YELLOW / RED approval gate, sector-specific weighting, and pattern alerts.\nThe key architectural decision: the numbers never come from the model. An LLM (Anthropic Claude or a local Ollama model) is used only to structure interview answers and to write narratives; every score, gate, and flag is computed by an engine with zero AI dependencies, so the same input always yields the same output. An optional RAG layer (ChromaDB) injects the relevant regulatory text into prompts, which measurably improves the quality of local models.\nThe framework is conservative by design: a skipped criterion still weighs on its dimension, a not-applicable criterion leaves the denominator entirely, and a partial assessment never produces a definitive gate.\nFeatures Deterministic scoring engine — gates, weighted sector profiles, pattern alerts, and normative flags, fully reproducible Pluggable LLM layer — Claude via the Anthropic API or Ollama for fully local, on-premise runs; adding a provider is one class and one line RAG over regulatory sources — EU AI Act and NIST AI RMF indexed in ChromaDB, with incremental re-indexing based on document checksums Interactive and batch modes — guided interview, JSON batch scoring, and partial re-assessment of selected dimensions Markdown reporting — scorecard, per-dimension criteria tables, executive summary, remediation plan, and regulatory appendix Traceability — every session records the LLM provider, the model, and the AIPAF version that produced it Jupyter notebook for exploratory assessment alongside the CLI Privacy by choice of provider Running with --llm ollama keeps the entire assessment — interview answers and RAG excerpts — on the local machine. This matters when the project under evaluation is itself confidential.\nRepository 🔗 GitHub - AIPAF\nCode released under MIT; framework documents under CC BY 4.0.\nAIPAF is a methodological support tool. Its scores and gates do not constitute legal advice, an AI Act conformity assessment, or an ISO/IEC 42001 certification.\nSkills Python (3.11+) Typer, Pydantic Anthropic API / Claude Ollama (local LLM) ChromaDB / RAG AI Governance (NIST AI RMF, ISO/IEC 42001, EU AI Act) pytest, ruff, GitHub Actions Last updated: August 2026\n","permalink":"https://www.lorenzolombardi.it/projects/aipaf/","summary":"Python CLI and methodology for structured assessment of enterprise AI projects, grounded in NIST AI RMF 1.0, ISO/IEC 42001:2023, and the EU AI Act","title":"AIPAF — AI Project Assessment Framework"},{"content":"Description A collection of development exercises and solutions sourced from platforms like HackerRank, written in multiple programming languages. A personal practice ground for keeping coding skills sharp across different scripting and programming environments.\nRepository 🔗 GitHub - Coding Test\nSkills Python (3.7+) Bash Script PowerShell Last updated: January 2026\n","permalink":"https://www.lorenzolombardi.it/projects/coding-test/","summary":"A collection of coding exercises and solutions in Python, Bash, and PowerShell","title":"Coding Test"},{"content":"Description A collection of Python and Bash automation scripts for Informatica products, including PowerCenter, Enterprise Data Catalog (EDC), Axon, Data Quality, and Informatica Intelligent Data Management Cloud (IDMC).\nRepository 🔗 GitHub - Informatica Automation Examples\nSkills Python Informatica PowerCenter Informatica IDMC REST API Automation Scripting Last updated: January 2026\n","permalink":"https://www.lorenzolombardi.it/projects/infa-automation-examples/","summary":"A collection of Python and Bash automation scripts for Informatica products","title":"Informatica Automation Examples"},{"content":"Description LyrixGram is a Telegram bot that allows you to search for song lyrics using the musiXmatch API. You can search by song title, artist name, or even lyrics fragments. Additionally, the bot provides a \u0026ldquo;lucky\u0026rdquo; feature that randomly selects a song for you.\nRepository 🔗 GitHub - LyrixGram\nSkills Python Telegram Bot API API Integration Last updated: January 2026\n","permalink":"https://www.lorenzolombardi.it/projects/lyrixgram/","summary":"Telegram bot for searching song lyrics","title":"LyrixGram"},{"content":"Lorenzo Lombardi Principal Data Architect | AI-Enhanced Data Governance \u0026amp; Lineage | Enterprise Solutions Architect at NTT DATA Italia, with over 20 years of experience in Enterprise Data Management — primarily in Banking and Insurance.\nTwo decades ago, when dot matrix printers and fax machines were still widespread, I decided to pursue a career in IT, a constantly evolving sector with unlimited potential for growth and transformation.\nThroughout my career, I have held various roles, including IT Manager, Head of Deliverability, and Product Manager. These experiences have taught me the importance of using the past as a guide without letting it become a constraint. It\u0026rsquo;s essential to remain open to new ideas and not get trapped in old certainties.\nI\u0026rsquo;ve also learned that titles describe roles, but outcomes are always the result of a crew. The helmsperson sets the course, the bowman makes it possible: every role on the boat matters, and a single wrong move puts everyone in the water.\nWhat I Do Today At NTT DATA, I lead data governance initiatives and develop complex architectural solutions for enterprise clients within the Data Intelligence \u0026amp; AI team. I also oversee the Informatica Competence Center, established as the central authority on Informatica implementations and best practices across the organization.\nOutside of client work, I am the author and maintainer of AIPAF — AI Project Assessment Framework, a structured, modular approach to evaluating enterprise AI projects grounded in NIST AI RMF 1.0, ISO/IEC 42001:2023, and the EU AI Act. The reference implementation — a Python CLI with a deterministic scoring engine and an optional RAG layer — is open source on GitHub under the MIT license.\nSpecializations Data Governance — Frameworks, policies, and processes for enterprise data management; regulatory compliance (BCBS 239 / RDA\u0026amp;RR, GDPR) Data Architecture — Design of scalable enterprise solutions for Banking and Insurance Informatica Suite — PowerCenter, Data Quality (IDQ), Enterprise Data Catalog (EDC), Axon, IDMC, Dynamic Data Masking (DDM), and PowerCenter-to-CDI migrations Data Lineage \u0026amp; Metadata — Solutions for compliance, data trust, and AI-enhanced discovery Agentic AI — Multi-agent architectures, Model Context Protocol (MCP) server design, LLM integration patterns, tool use / function calling Generative AI for Data — Retrieval-Augmented Generation (RAG), ML for anomaly detection and predictive data quality, LLM integration in governance workflows AI Governance — Structured assessment of AI projects against NIST AI RMF, ISO/IEC 42001, and the EU AI Act Integration \u0026amp; Automation — Python and Bash scripting to integrate, monitor, and extract information from enterprise platforms Certifications AI \u0026amp; Generative AI\nOpenAI Technical Practitioner - PartnerU AI Agentic Applications Masterclass — Data Masters Claude Code in Action — Anthropic Introduction to Model Context Protocol — Anthropic Model Context Protocol: Advanced Topics — Anthropic GPTS-CLAIRE AI Foundation Certification — Informatica Building AI — University of Helsinki Elements of AI — University of Helsinki Data Governance \u0026amp; Informatica\nData Governance Implementation Practitioner — Informatica Cloud Data Integration for PowerCenter Developers, Foundation Level — Informatica IT Service Management\nITIL 4 Foundation Certificate in IT Service Management — PeopleCert / AXELOS Languages 🇮🇹 Italian (native) 🇬🇧 English (professional — B2) 🇫🇷 French (basic) For a detailed overview of my career history and professional background, visit my LinkedIn profile. I also write on Substack.\nThis is my personal space where I share projects, technical articles, and useful resources in the field of data management.\n","permalink":"https://www.lorenzolombardi.it/about/","summary":"\u003ch2 id=\"lorenzo-lombardi\"\u003eLorenzo Lombardi\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003ePrincipal Data Architect | AI-Enhanced Data Governance \u0026amp; Lineage | Enterprise Solutions Architect\u003c/strong\u003e at NTT DATA Italia, with over 20 years of experience in Enterprise Data Management — primarily in Banking and Insurance.\u003c/p\u003e\n\u003cp\u003eTwo decades ago, when dot matrix printers and fax machines were still widespread, I decided to pursue a career in IT, a constantly evolving sector with unlimited potential for growth and transformation.\u003c/p\u003e\n\u003cp\u003eThroughout my career, I have held various roles, including IT Manager, Head of Deliverability, and Product Manager. These experiences have taught me the importance of using the past as a guide without letting it become a constraint. It\u0026rsquo;s essential to remain open to new ideas and not get trapped in old certainties.\u003c/p\u003e","title":"About Me"}]