Amy X. Chen

A progress-resilience lens on urban planning today
@readloadbearing on Substack

[Amy X. Chen @amyxchen_]

Two nights before Matariki, Auckland looks like this.

Not a render. Every light is a record — 61,228 real buildings, LiDAR terrain, live transit. Even the stars rise on real sidereal time.

One HTML file. Map + code drop Friday. 🔊

[Amy X. Chen @amyxchen_]

Check out my latest article: A working archive of every Nobel Memorial Prize in Economic Sciences, 1969–2025 https://t.co/rQhb1eHl0z via @LinkedIn

I built something I wish I'd had before I declared my economics major.

My undergrad arc taught me a lot, mostly one course at a time. Financial econ: CAPM and pricing risk. Political economy: Coase's theorem. Market design: deferred-acceptance algorithms, which is how MIT places students into dorms, how the medical-residency match works, how kidney-exchange chains save lives, how the FCC auctions spectrum. International trade: Krugman's textbook and the new trade theory. Social choice: working through the proof of Arrow's impossibility theorem on a problem set. Environmental economics: cap-and-trade.

What I never had was a single place where I could see how these fit together — the through-lines from Coase to cap-and-trade to the EU's emissions market; from Vickrey to spectrum to Google ads; from Arrow to Sen to the UN Human Development Index. And there were entire branches my curriculum never touched: Solow growth, rational-expectations macro, behavioral economics, causal inference, climate-integrated assessment, institutions.

So I built the thing I'd wanted as a sophomore. A working archive of every Nobel Memorial Prize in Economic Sciences from 1969 to 2025. All 57 prizes, all 96 laureates, each with a plain-English explanation, a specialist's note, a section on how the work is used today, and links to the original papers.

[Amy X. Chen @amyxchen_]

My brain at 2am: these could be good college bowl quiz hints.
- Ode and ODE
- Jacobian and The Jacobian

Ode: An ode is a formal, often ceremonious lyric poem that addresses, praises, or glorifies a person, object, place, or abstract idea.

ODE: Ordinary Differential Equations (ODEs) are equations involving functions of a single independent variable and their derivatives, used to model dynamic systems by relating rates of change to current states. They are essential in physics, engineering, and biology for predicting phenomena like motion, population growth, and electrical circuit behavior.

Jacobian: Jane Jacobs (1916–2006) was an influential activist and theorist who revolutionized urban planning by championing organic, high-density, mixed-use neighborhoods over top-down "urban renewal". Her landmark book, The Death and Life of Great American Cities (1961), argued for "eyes on the street" for safety, diverse functions, and community-based development. I use Jacobian to refer to Jacobs’ influential ideas.

The Jacobian: The Jacobian matrix is a matrix of all first-order partial derivatives of a vector-valued function, representing its best local linear approximation at a given point.

[Amy X. Chen @amyxchen_]

@_Shark_byte Love the site! One suggestion is to use a different font/design language of the site that is not super Claude coded; and I am a little confused about what the colors of the buildings represent, as neighborhoods are also color coded, so can make this more intentional.

[Amy X. Chen @amyxchen_]

Check out my latest article: How I built an editorial website about the Bronx River in three hours https://t.co/24HcUjmmM7 via @LinkedIn

[Yohan @yohaniddawela]

Google trained an AI to predict your neighbourhood's income by counting the coffee shops, bus stops, and high-rises on a map. Nobody told it what income was.

The model is called S2Vec, published this month by Google Research as part of their Earth AI initiative. It takes the built environment (every building, road, park, and business in an area) and converts it into a layered image. Three coffee shops and one park in a grid cell become pixel values. The AI then reads that image the same way a computer vision model reads a photograph.

The training method is the part that matters. S2Vec uses masked autoencoding: you show the model a patch of a city with chunks missing, and it learns to fill in the gaps. Show it a cluster of high-rise apartments next to a subway station, mask out a section, and it predicts a grocery store belongs there.

Do that millions of times across the globe and the model learns the deep spatial grammar of how cities organise themselves. No human ever labels a region as "financial district" or "suburban residential." The model figures out those groupings on its own from the geometry of what's built where.

The output is an embedding, a string of numbers that acts as a mathematical fingerprint for any location on Earth. Feed those embeddings into a prediction task and S2Vec can estimate population density, median income, and carbon emissions for regions it has never seen before.

On zero-shot geographic extrapolation (predicting for regions entirely absent from training data) S2Vec was typically the best-performing individual model.

It matched or beat satellite imagery baselines like RS-MaMMUT and outperformed GEOCLIP on socioeconomic prediction. The best results came from combining S2Vec with satellite image embeddings. Built environment data alone couldn't capture vegetation, terrain, or transportation patterns well enough for environmental tasks like tree cover and elevation. But fused together, the two modalities outperformed everything else.

The standard approach to geospatial ML has been hand-crafting indicators for every new problem. Predicting air quality meant building a bespoke feature set. Estimating housing prices meant building another one. S2Vec replaces that with a single general-purpose representation that transfers across tasks.

The training data is map features, not satellite pixels.

That distinction is pretty important to understand. It means: map data updates faster, costs less to process, and covers built infrastructure at a resolution satellite imagery can't always match.

A satellite sees rooftops. S2Vec knows there are three cafes, a pharmacy, and a bus stop underneath them.

Google's broader Earth AI pipeline now has three foundation models working in parallel.

1. PDFM for population dynamics.
2. RS-MaMMUT for satellite imagery.
3. S2Vec for the built environment.

Stack them and you get a system that can read a neighbourhood the way a local understands it.

[Kat ⊷ the Poet Engineer @poetengineer__]

trying to use topological data analysis to map the shape of my x bookmarks through mapper + embedding extraction and generated 3 views:
- density: where attention keeps gravitating
- pca: the dominant axes of variation
- centroid: center vs edge (typical -> outlier)

[Amy X. Chen @amyxchen_]

I sometimes write about culture, and last night, while studying for ODE and listening to my Spotify playlist, I was reminded of the fond memories I have for some 80s and 90s Cantonese songs. It’s an interesting question how I, as a Gen Z Chinese kid, can resonate so deeply with that specific era of Cantonese music. I think part of the answer lies in history: from the rapid urban to rural adoption of color television in mainland China to the way TV sets became shared neighborhood amenities, shaping collective experiences and elevating certain singers into the hall of fame. The golden age of Cantopop feels closely tied to this mix of cultural amalgamation- many Cantopop were adapted (改编 - gǎibiān) from Japanese originals- and technological change.
https://t.co/9QgvJTlFfL

[Amy X. Chen @amyxchen_]

How to see a city series episode 1: scale & friction

[Amy X. Chen @amyxchen_]

The City (1939 Documentary) https://t.co/0OSU4FZNFX via @YouTube

[Amy X. Chen @amyxchen_]

When I become a city planner I will travel to work on urban design and development projects, help governments build open data platforms and ecosystems, lead ML/remote sensing research on land use & mobility, and advocate for policies that improve economic and social outcomes.

[Amy X. Chen @amyxchen_]

@DAIEvolutionHub Need

[Kshitij Mishra | AI & Tech @DAIEvolutionHub]

Access 15 FREE Courses

🔥 Microsoft Excel
🔥 Advanced Excel
🔥 Probability
🔥 Statistics
🔥 Mathematics
🔥 Tableau
🔥 PowerBI
🔥 SQL
🔥 SQL + Tableau
🔥 Python
🔥 Python Programmer Bootcamp
🔥 Git and GitHub

Just do - Follow ( So I can DM you )
Like & Retweet
Comment " Need "

[Amy X. Chen @amyxchen_]

@RodmanAi LLM

[Amy X. Chen @amyxchen_]

@Alexcoderzx Claude

[Amy X. Chen @amyxchen_]

**You cannot understand urban planning without understanding path dependence**. This is the fall line in northeast America. A geological boundary where rivers drop from the older Piedmont to the younger Coastal Plain, leading to cities like Philadelphia, Richmond, and Augusta.

[Amy X. Chen @amyxchen_]

Started this account to complement my substack and share more personal reflections on startup progress. Forcing myself to tweet everyday. Would love to build a community with people interested in urban tech, real estate, infrastructure, navigation, education and lifestyle.

[Amy X. Chen @amyxchen_]

@NainsiDwiv50980 AI

[Nainsi Dwivedi @NainsiDwiv50980]

All Paid Courses (Free for First 4000 People)

🎯 LLM Mastery (GPT, Claude, Grok)
🎯 AI Agent Building Masterclass
🎯 Advanced Prompt Engineering
🎯 AI Automation with Zapier & Make
🎯 No-Code AI SaaS Building
🎯 RAG Systems & Vector Databases
🎯 LangChain & Agent Frameworks
🎯 AI Content Monetization
🎯 AI for Freelancers
🎯 Chatbot Development
🎯 AI App Development
🎯 Generative AI for Business
🎯 AI Workflow Systems
🎯 And more...

FREE for 48hrs only

To get Simply:

1) Follow me [ MUST! ] for DM access
2) Like & RETWEET
3) Comment "AI " to grab your copies

[Amy X. Chen @amyxchen_]

@yourtechgirl24 SEND