A Methodology Without a Tool
A vision without a vessel.
The client is a researcher and educator whose forensic methodology for analyzing complex societal problems is unlike anything in the market. The approach maps entangled causal relationships across time, discipline, and scale. Principal factors, contributing causes, cross-connections, epistemic confidence levels, all in a visualization so mathematically demanding that the first question wasn't whether TDG could build it. It was whether it could be built at all.
The answer was yes, and that opened the door for even more important questions.
There's no spec sheet for a tool that doesn't exist yet. There's a vision, a methodology, a set of needs the client can articulate, and a larger set she can't, because you can't know what a tool needs until you've used it on real work with real clients.
Ship the vision. Stay embedded. Let the work reveal what's next.
The engagement moved through three distinct phases. Each one changed what the tool fundamentally was.
We can build this vision, but this is just step one. The best part will be a year from now, when the product has evolved so far you won't recognize what we delivered on day one.
A second axis
Most AI products stop at one axis: different agents doing different jobs inside a shared context. TDG is building a second axis on top of it, and this is the part that catapults the client's practice ahead of the broader research field.
A research project isn't the same kind of work on day one as it is on day ninety. Early on, the researcher is exploring broadly and the map barely exists. By the final stage, the map is the deliverable and everything else is reference material. Right now, every AI harness treats those two moments identically. TDG's thesis is that it shouldn't.
That work is already underway. The harness weights toward the structured map during Analysis, the stage where the researcher's own rulings carry the most weight, and both Ask and Build operate live within it. Sensemaking tuning, which teaches the harness to lean on raw source material and the emerging map together while the argument is still taking shape, is the piece still in active development. Together, the harness starts to feel less like a tool and more like a research partner that knows what stage of thinking it's in.
A central nervous system for the practice
The full vision is a central nervous system for the entire research practice, where diverse inputs — papers, ethnographies, hard data — feed into the map. The map produces findings. The harness carries what it's learned into the next project. Every engagement makes the next one smarter.
This is what it means to build a genuinely AI-native organization: one where AI is woven into how the work is structured from the ground up, rather than added on top of existing workflows.
Where things stand
- Deterministic platform shipped and in active use across a full research engagement, start to finish.
- The researcher has moved into her second project on the same platform without disruption.
- Ask agent live, reasoning directly from the project's own knowledge base.
- Build agent live, proposing factor additions and new interconnections for the researcher to accept or reject.
- Analysis-stage lifecycle tuning live, weighting the harness toward the structured map when the researcher's rulings matter most.
- Sensemaking-stage lifecycle tuning in active development.
- Challenge agent and the remaining lifecycle stages (Discovery, Presentation) queued as the next architectural step.
Explore More Transformations
See how we've helped other mission-driven organizations achieve breakthrough results.