the full chartales workflow:
1. drop file
2. AI surfaces questions
3. pick one or type your own
4. chart + insight together
5. refine in plain language
6. export PNG · SVG · PDF
no code. no data team.
launching May 1st
chartales.com/join-waitlist
#30DayChartChallenge #dataviz #buildinpublic
built this for everyone who's ever had something important to say with data but couldn't get the tools to cooperate
upload your file · ask a question · get a chart + story
ready to share minutes not hours
launching May 1st
chartales.com/join-waitlist
#30DayChartChallenge #dataviz #buildinpublic
This has been a fun exploration using AI as a thinking aid, not a black box.
Still early and still evolving, but I wanted to share where this is heading.
Would this be something that could help your workflow?
I’d love to hear your thoughts and any feedback you might have.
I repeat this with three different datasets, each with a different question, to see how well this flow holds up across contexts.
This isn’t about replacing analysis.
It’s about reducing the friction between having data and starting to understand it.
In this short demo, I’m testing a simple workflow I’m calling CSV to Insights.
What’s happening in the video:
– I upload a CSV
– I ask one short question about the data, or leave it blank (letting the AI decide)
– The system generates an insight
– And produces a chart to anchor that insight
Hello data explorers,
I previously shared an experiment called AI CSV to Insights. I’ve since resumed work and have been quietly continuing this small exploration around a very familiar problem:
You open a CSV… and you don’t know where to start.
#CebuEarthquake #Aftershocks #PHIVOLCS #DataViz #GIS #DisasterPreparedness
Tools: Visualization created in QGIS, annotations in After Effects, Python scripting assisted by ChatGPT
We see where the aftershocks were happening, how strong they were, and how the sequence evolved.
Data source: PHIVOLCS Earthquake Information (Sept 30–Oct 2)
Note: This is an explanatory visualization, not a real-time alert. For safety guidance, always follow PHIVOLCS and your local authorities.
The first frames highlight the main shock, followed by notable M4.8 and M4.5 aftershocks clustered offshore. Over time, the animation shows how activity concentrated along the same fault.
Why visualize this? Earthquakes can feel chaotic in the moment. By mapping the data, patterns become clearer.
This animated bubble map shows the aftershock sequence from September 30 to October 2. Each circle represents one quake, with its size scaled to the magnitude. Each frame of the video corresponds to about 30 to 40 minutes, allowing us to see how seismic activity unfolded.
Cebu Aftershocks, visualized (Sept 30–Oct 2)
On the evening of September 30 at 9:59 PM, a magnitude 6.9 earthquake struck offshore northeast of Bogo City, Cebu. In the days that followed, hundreds of aftershocks rippled across the area, with consecutive ones recorded up to magnitude 4.8 and 4.5.
Flood risks are not just numbers—they’re lived realities. Data can help us see what’s at stake, and where accountability matters most.
#Philippines #FloodControl #DataViz #GIS #Corruption
🛠️ Behind the build:
Data: Global Flood Database + PSGC (Philippine boundaries)
Extraction: Google Earth Engine + AI-assisted scripting
Processing: QGIS + grid redistribution for population stats
Visualization: Observable Framework + DeckGL HexagonLayer
- In Mindanao, recurrent floods in Zamboanga del Sur, Davao del Sur, and Misamis Oriental highlight growing risks outside Luzon.
📊 Key Insights:
- Bulacan and Pampanga were among the most flood-prone provinces with around 6.5M affected in affected areas. Interestingly, Bulacan also had the highest number and cost of flood control projects—where many substandard or ghost projects were also later discovered.
🔍 How to read the visualization:
Color of the hexagons = frequency of flooding (the redder, the more frequent)
Height of the hexagons = number of people exposed within that grid
Hover to highlight provinces for details
Before corruption cases even surfaced a month ago, I built an exploratory visualization prototype to better understand flood risks across ASEAN. I mapped flood-prone areas and the number of affected people at a granular level.
Hello data explorers,
Flood control failures and corruption scandals take center stage in the Philippines. Budget for flood control projects have recently been reported down to zero according to Pres. Marcos for 2026.
Here's the catch, it is using an actual data which was taken from Yale University's Geographically based Economy Data: gecon.yale.edu
Would you want to have a postcard of it? Let me know if this is mini-project you are keen to explore further.
- The area of the hash patterns represent the actual cropland of the surrounding rural areas.
- The color theme shows whether you are situated within a tropical, subtropical and other vegetation types.
Here's my attempt to visualize someone else's city within a 1x1 degree of the world as a mini digital township.
Every visual element is encoded within 1x1 degree with:
- The size of your town or city represents by the actual population.
What if your town or city is visualized as a miniature isometric village? Can you imagine what your city in a postcard would look like?
.
.
.
I was trying to organize my files and stumbled upon an archived project.
Like sculpting, you take raw materials, squish them together, trim the excesses and voila!
This is not how usual design and development process actually works, but for the sake of figuring out something new, let's be messy. XD
#deckgl #dataexploration #innovation
And the exploration alone is already a joyful process XD.
Speaking about mess, I basically dumped all potential data points I want to use for a potential 3d mapsploration, and potentially carve this to the point where it can be used for accessing key insights and make it even usable.
Mess, mess, mess...
I don't have a proper context to share yet and nothing fancy, but I'm playing around on Deck GL's capabilities to handle hundreds of thousands or even millions of data points. It's something that I've been longing to explore in a long time.
Let data guide your next trip (or analysis)
Explore the dashboard + map → traveltrendsph.vercel.app
#DataViz #PhilippineTourism #Observable #OpenData #GIS #TravelTrends #TourismInsights
📊 What I discovered
🇵🇭 95M recorded visits in 2019, 2021, and 2023
📉 Foreign tourism dropped 35% from 2019 to 2023
🏙️ NCR saw a surprising +240% surge in local tourism
👀 Many lesser-known places are gaining fast — are you watching them?
🛠️ How I made it
Cleaned shapefiles and tabular data from government PDF reports. Mapped and visualized using Observable Framework, Plot, QGIS, and Mapshaper
