Pressure Atlas laptop screen in a cinematic night studio.
Pressure Atlas on a laptop with eyeglasses at a studio desk at night.

Pressure Atlas

Designing the long view of finance and politics

The Observatory’s research, translated into a story readers can examine.

What do all these stories amount to?

Finance and politics arrive as a stream of separate stories. Following the news can tell us what happened without explaining how much the bigger picture has changed.

The Observatory began with a desire to follow those connections over time. Pressure Atlas gives its research a consistent reading experience, where new evidence can strengthen an interpretation or change its direction.

Project snapshot

Project

The Observatory / Pressure Atlas. An independent research and publishing system for examining finance and politics.

My role

Concept and design direction across the research system and reading experience, developed with substantial AI collaboration.

Design problem

Connect fragmented information into an evolving argument that readers can inspect and question.

Delivered

A functioning web briefing with evidence checks, documented handoffs and recovery of an interrupted release.

Designing the system behind the screen

I conceived the Observatory and directed its development over months of iteration with several AI models. The design work included deciding how research becomes a publishable argument, then shaping the software that lets readers examine it.

  • Qualify evidence before it supports a claim.
  • Keep contrary evidence beside the interpretation it limits.
  • Carry an approved briefing through product checks before release.
  • Record completed work and unresolved issues for the next session.
 

A working thesis about connected pressure

The research follows pressure through four stages. Each connection is an argument to examine, with evidence that can support or interrupt the sequence.
 
01

Price pressure

Persistent inflation can keep everyday costs high and limit the relief households feel.

02

Fed constraint

Inflation can restrict policymakers’ room to ease. The assessment asks how much flexibility remains.

03

Credit strain

Expensive financing reaches households and businesses. Strain in one channel needs to be distinguished from a broader loss of access to credit.

04

Funding confidence

The research examines whether access to funding is weakening. Pressure in an earlier stage cannot establish the whole sequence.

 
 

Follow the argument

Move from the connected pressures to the evidence, then examine what would change the assessment.
Close view of the Price Pressure stage at 94 percent.
Causal Atlas screen detail with Price Pressure at 94 percent.

01
Read the connections

Causal Atlas places each stage beside its resistance and consequences, keeping the reasoning within reach.
Close view of the Evidence Field and a selected source receipt.
Evidence Field screen with source receipts beside the assessment.

02
Inspect the evidence

The Evidence Field lets readers examine the source material behind a claim and the counterpressure that qualifies it.
Close view of Watch Floor conditions and their consequences.
Watch Floor screen showing conditions that would change the assessment.

03
Watch for a change

Watch Floor names observable conditions and explains how they would strengthen or weaken the assessment.
 
 
86% BLEED PATH photographed at macro range, with visible RGB screen pixels and shallow focus.

Confidence needs a definition

86% describes confidence that the pressure regime is active. It is not a probability of collapse.
Credit Strain screen detail showing evidence that limits the claim.

The evidence that limits the claim

Continued lending and functioning credit markets matter alongside signs of strain. Counterpressure changes how far the conclusion can go.
Screen pixel detail of a Watch Floor threshold and its consequence.

The condition that changes the read

A Watch Floor condition connects an observable threshold to its consequence for the assessment. Readers can see what the research is waiting to learn.
 
 
 

Built for the days that do not go to plan

The system was refined through months of use, including runs that could not proceed. It needed a clear response when evidence did not qualify and when a product check interrupted a release.
 

Held

September 28, 2026

Insufficient eligible fresh evidence. The previous briefing stayed available.

Resumed

October 8, 2026

A mobile layout defect was repaired. The unfinished release continued with its evidence intact.

 

Continuity is part of the product

The next session inherits the work

The handoff records the state of the run, including completed work and the checks still outstanding. After the October 8 repair, the same briefing reached production without repeating its research.

For readers, continuity means the published argument stays available as the next assessment develops.
 
 
 

From research to a checked briefing

The workflow connects research to publication through explicit checks. Each stage has a purpose, including a handoff that closes the run or preserves an unfinished release.
 

One connected publishing workflow

 
  1. 01

    Research

    Gather the evidence

  2. 02

    Assessment

    Qualify what can support a claim

  3. 03

    Narrative

    Develop a connected interpretation

  4. 04

    Web briefing

    Present the argument for examination

  5. 05

    Release checks

    Publish when the evidence and product pass

  6. 06

    Handoff

    Record the state for the next session

 

A run starts manually. Publication depends on the checks.

 

The product includes the decision to stop.
Charles HurstDesign principle
  1. 01 · Qualify sources
  2. 02 · Withhold publication
  3. 03 · Close the run

Previous briefing stays available.
Insufficient fresh evidenceSeptember 28, 2026
  1. 01 · Record the issue
  2. 02 · Repair and resume
  3. 03 · Verify publication

The unfinished run keeps its evidence.
Interrupted releaseOctober 8, 2026
 

Design direction across the whole system

 

Concept and direction

The project began with my own need for a longer view. I directed its development around a reader’s ability to follow an argument and understand its limits.

Working with AI

Several AI models contributed to the research and software implementation. The collaboration made it possible to develop the system through repeated use and revision.

Design beyond the interface

The work reached into publication behavior and recovery. Deciding what readers could rely on shaped the process as much as the screen.

 

Bring me the difficult problem

For teams working with complex information, I offer design direction grounded in how the underlying system behaves. I’m open to project engagements and senior design leadership roles.