Data Analyst | Power BI, SQL, Python | QA & Manual Testing | AI Specialist

Python really clicked for me when I was selected as one of only three students our faculty picked to join their data science project for the local power utility provider of Iloilo City. We were handed raw, messy operational datasets and had to make them usable. I built a cleaning and transformation pipeline from scratch, including feature engineering like calculating turnaround time and writing a script to parse District and Barangay out of an inconsistent, messy address field. Seeing it turn chaos into something the faculty could actually deliver to the client is what hooked me. Python was also the main programming language I used to build the machine learning model for our thesis, an LSTM model for weather prediction, so I've worked with it well beyond data cleaning into actual model development. Since then I've used Python for ETL pipelines, exploratory data analysis, and automation scripting, working mostly with pandas and numpy. I've also used it to build backend logic for AI systems, so I'm comfortable with it both as a data tool and a development one.

SQL is the tool I reach for without thinking. On data projects I've used it to extract, join, filter, and aggregate across multiple tables, not just for one-off queries but as a repeatable step inside a larger pipeline alongside Python and BI tools. I'm comfortable in relational databases like PostgreSQL and in cloud environments. I learned early that a well-structured query saves hours downstream, so I write SQL with the next person, or the next stage of the pipeline, in mind.

I did manual testing during my internship on real, deployed systems used by very different users. On the school platform I tested across admin, teacher, student, and parent roles, catching data integration issues like a content configuration screen that failed to load existing items when historical records existed. On the multi-agent AI marketing system I tested the full auth and onboarding flow end to end, from signup and login to the multi-step onboarding and the AI conversational intake that generates a GTM playbook. Testing flows that real users depend on taught me to test how people actually behave, not just the happy path, and to log everything systematically so nothing slips.

My strongest analytics experience came from being chosen, with two other students, for our faculty's data science project with the local power utility provider of Iloilo City. I ran exploratory analysis on large, real operational datasets and did hands-on feature engineering, deriving turnaround time metrics and extracting structured location data (District and Barangay) from a messy free-text address field so it could actually be analyzed. The job was finding the patterns and anomalies hiding in data nobody had cleaned yet, then shaping it into something the team could deliver. My thesis pushed this further, applying an LSTM deep learning model to time-series weather data, which grounded me in statistical thinking and model evaluation, not just charting.

Excel and Google Sheets are where I appreciated structured data the most, especially during QA work where I ran the entire bug-tracking and test-documentation system out of a spreadsheet. Every test case had its category, scenario, steps, priority, status, owner, and detailed notes, all laid out so the dev team could scan and act without back-and-forth. That experience taught me that a spreadsheet is only as useful as it is readable. I'm comfortable with pivot tables, VLOOKUP, INDEX-MATCH, conditional formatting, and building quick dashboards, and I know how to keep data organized for whoever needs it.

During my internship I tested real, deployed software, not practice apps, across three live systems. My focus was making sure features worked the same across different browsers and devices, and that things held up in real use, not just the happy path. I logged every issue with steps to reproduce, priority, and status, mostly in Excel and Google Sheets, and we also tried Trello and Jira a bit. What I think sets me apart is I don't stop at what's broken: on the school platform I noticed the dashboard loaded all student data at once and suggested paginating it so it loads faster. I test to make the product better, not just to file bugs.

I did this during my internship, and my QA strength is catching the subtle problems that pass a quick look. The one I'm proudest of: the same student showing different grades between the Progress Report and Student Directory modules, because the two pulled from different data sources. That kind of silent inconsistency quietly destroys trust in a system, and finding it is what made QA matter to me. I write reports that state the issue, the expected behavior, and a severity rating clearly enough that a developer can fix it without a single follow-up question.

What sharpened my Power BI skills was building reporting dashboards meant for people who don't live in spreadsheets, like admins and operations staff who just need answers fast. I work with DAX for custom measures, connect multiple data sources, and lay out reports so a non-technical user can find what they need without asking me.

My AI experience is genuinely hands-on, both building with AI and building AI systems. I co-developed a live multi-agent marketing system where autonomous agents each own a distinct role and work together as one pipeline, and I helped build a full school management system using AI app builders like Base44. I work daily with AI tools like ChatGPT, Claude Code, and Codex as part of how I actually develop and ship software. Earlier, my LSTM weather-forecasting thesis taught me the fundamentals of how a neural network is trained, evaluated, and improved. I'm comfortable with AI not just as a tool I use, but as something I build systems around.