A Simple Methodology: Discover → Analyze → Challenge → Validate AI can be an incredibly useful strategic partner for a technology startup, but there is an important lesson from this conversation with Gemini: don’t ask AI for the answer before giving it enough information to develop a good answer. A better approach is to use […]
Gemini Builds Your Tech Startup Strategy—If You Ask the Right Questions Read More »
In this post, we will explore how to use SQL window functions within the Gold layer to solve two common needs: calculating granular aggregates and generating synthetic test data for healthcare scenarios. Modern data engineering has converged on the Medallion Architecture as the gold standard for managing scalable data lakes. By organizing data into Bronze
Introduction Many organizations face the problem of choosing optimal locations for facilities — whether warehouses, travel hubs, corporate headquarters, or distribution centers. The goal is to minimize travel time, improve logistics efficiency, or maximize accessibility. This is a general location-optimization problem, applicable in many contexts: While the domain changes, the core problem remains: finding the
Solve Network Optimization Problems with Python, NumPy, Scikit-learn, and OSRM Read More »
Executive Summary Master Data Management (MDM) is the discipline that enables organizations to maintain accurate, consistent, and trusted domain-specific master entities—customers, products, distributors, patients, and more—across all systems. It ensures enterprise operations and analytics are based on a single source of truth. Historically, MDM has been slow and resource-intensive, creating latency between data capture and
AI Acceleration of Master Data Management Read More »
Understanding Context and RAG Before talking about RAG, it’s important to understand context in large language models (LLMs).When you send a prompt, the LLM first breaks it into tokens — small units of text, roughly 3–4 characters of English on average. These tokens, along with any chat history and documents you provide, make up the
Why RAG is not just for “teaching” LLMs Read More »
Hi, I’m Daniel Acevedo—founder and principal author and consultant at IT and Management. I’m an IT strategist who enjoys making sense of messy systems, guiding teams toward clarity, and building bridges between technology and business. Over the years, I’ve worked in IT across industries, but what drives me is pretty simple: helping people solve problems
Training a neural network is how we teach machines to recognize patterns — like identifying faces, translating languages, or recommending products. It’s the core of how modern AI systems learn from data and improve over time. Learning how to train a network unlocks powerful tools for solving real-world problems, and it’s more approachable than you
How to Train Your Neural Network Read More »
In today’s world of AI innovation, having the power to run large language models (LLMs) locally is becoming increasingly valuable. Whether you’re a researcher, developer, or AI enthusiast, bringing your AI work home means taking control of your environment, data privacy, and workflow efficiency without relying on cloud services. With advancements in GPU technology and
Bringing Your AI Work Home — Setting Up for Local LLMs Read More »
The field of AI Agent Engineering is exploding with opportunity. The good news is that the tools and knowledge you need are now available to everyone, regardless of background. This post gives you a roadmap and curated resources to help you launch your journey with focus and momentum. AI has been fully democratized. This means
Prerequisites for AI Agent Engineer Read More »
If you’ve ever needed to label your data for downstream AI, you know how tedious and error-prone manual tagging can be. Fortunately, with the help of Python and NLP tools like spaCy, you can automate this process and even train a model that understands your domain. In this workshop, you’ll build a system that classifies
As a data engineer navigating cloud platforms and analytics ecosystems, I’ve worked extensively with both Databricks and Azure Synapse Analytics. While they often appear side-by-side in Azure environments, they come from different design philosophies and cater to slightly different needs—even when they seem to do many of the same things. Both platforms allow you to
This is the first in a series on understanding systems through the lens of deterministic and probabilistic design. We’ll start with the core distinction, then look at real-world examples—like MDM matching and network optimization—before exploring the more fluid world of probabilistic models and AI systems. In designing systems—whether they’re used for reporting, automation, machine learning,
Deterministic vs. Probabilistic Thinking: Tools for the Thoughtful Technologist Read More »
When One Field Tries to Do Too Much Across many organizations, the same pattern repeats. A single data field—perhaps introduced during a system rollout or legacy data migration—is intended to serve a focused purpose. Over time, however, teams start using it to represent different ideas. Marketing redefines the field to support campaigns. Sales uses it
The Hidden Cost of Overloaded Data Fields — And How Data Governance Saves the Day Read More »
When building cloud data pipelines with Azure Data Factory (ADF), you’re often presented with two paths: use the tried-and-true Copy Activity, or lean into the newer, powerful Mapping Data flows. Both are native ADF features and technically serve the same purpose—moving and transforming data—but the similarities end there. In our case, we started with Data
I worked many years with the NPI dataset. Transforming the data brought actionable insight. Imagine a conversation between an enterprise IT strategist, a healthcare IT consultant, and a healthcare administrator. The three sit around a conference table reviewing a dataset pulled from the National Plan and Provider Enumeration System (NPPES), full of NPIs, taxonomy codes,
Using SQL UNPIVOT to Unlock Healthcare Provider Insights with Unpivoted Taxonomy Data Read More »
What I Used to Think About Negotiation Many people—including myself in earlier years—view negotiation as a zero-sum game: if I win, you lose. This mindset defines distributive negotiations, where parties focus on a single issue (typically price), each trying to claim as much value as possible from a fixed pie. But there’s a better way.
Creating Value in Vendor Negotiations – Insurance Underwriting Example Read More »
In fact there is not a more suitable collection of problems than those found in IT. I’ll mention two and walk through the second – Data Quality. Use Six Sigma to squash bugs. Those are the pesky defects that litter software applications. Reducing defects is the very reason Sig Sigma exists. Look no further than
Six Sigma’s new home in IT Read More »
so you can report off of it. Note: I actually profile my Databases completely different than what is presented here, but this information might be useful to someone else. If you have Microsoft SSIS, then you can take advantage of the Data Profiling Task. It profiles your tables and columns and outputs an XML file.
Get that Free Profile Data Into Your Database Read More »
The factory floor was buzzing — but the numbers didn’t add up.A midsize manufacturer had ramped up production in anticipation of a seasonal spike, only to find themselves overstocked on one product line and chronically short on another. Overtime costs spiked. Storage overflowed. Shipping delays mounted. Managers were making decisions fast, but not always aligned
Operations Management Science for Growth – a Framework Read More »
I started out considering MDM, but then… I truly believe that the process is going to be the big differentiator in the Data Management space in IT. The technology will just aid the process. Actually, the process can commence before technology is ready and then technology can catch up easily. As I dug into this
My notes for Standing up a Data Management Program Read More »
Many organizations struggle with duplicate customer sites and accounts scattered across operational systems. These duplicates introduce confusion in invoicing, shipping, reporting, and analytics. The immediate need to resolve this data fragmentation often outpaces the availability of enterprise-wide tooling. This post outlines how we built a pragmatic, SQL-based deduplication framework to identify, cluster, and prepare duplicate
Dealing with Duplicates: A Practical SQL-Based Deduplication System Read More »
When designing real-world applications, I’ve found that simplicity is more than elegance—it’s strategic. Simple code is easier to secure, easier to maintain, and more scalable in the long run. Edsger W. Dijkstra, a renowned computer scientist, famously said: “Simplicity is prerequisite for reliability”. The KISS principle (“Keep It Simple, Stupid”) is a widely embraced design heuristic in
To qualify for the PMI Agile Certified Practitioner (PMI-ACP®) credential, I documented my work on several agile projects. As I reflected on these experiences, I realized how well they illustrated many of the core principles embedded in the PMI-ACP framework. Agile isn’t simply a methodology—it’s a mindset that values individuals and interactions, customer collaboration, responding
Three Projects Shaped My PMI-ACP Journey Read More »
It helps to create all types of indexes in our head when studying for any particular matter. This is why songs can be easy to memorize. Likewise, printed materials are often times easier to learn than electronic ones. I searched for a poster of the PMP knowledge areas and process groups. In particular, I wanted
PMP Knowledge Areas and Processes Poster Read More »
In the information age, competitive advantage comes from making business sense of vast amounts of data. Departments demand Business Intelligence for specific local needs that may not be offered by a central enterprise BI team. How do we begin to meet these needs in an orderly way? This post addresses this need with a framework
A Framework for Self-Service Business Intelligence Read More »
