Start Here¶

How does a first-time visitor self-identify, and which mental path leads them to real value?

flowchart TB
%% Styles (brModel Standard)
classDef i fill:#D3D3D3,stroke-width:0px,color:#000;
classDef s fill:#FFB3B3,stroke-width:0px,color:#000;
classDef p fill:#B3D9FF,stroke-width:0px,color:#000;
classDef r fill:#FFFFB3,stroke-width:0px,color:#000;
classDef o fill:#C1F0C1,stroke-width:0px,color:#000;

S_Visitor("👤 Visitor (YOU)"):::s
P_HowWorks("ℹ️ How this site works"):::p
P_Ident{"Choose your role"}:::p

S_Buyer("🧑‍💼 Buyer / operator"):::s
S_Eng("🧑‍💻 Engineer / architect"):::s
S_Read("🧑‍🔬 Reader / analyst"):::s

S_Visitor --> P_Ident
S_Visitor -. "need context" .-> P_HowWorks --> P_Ident
P_Ident --> S_Buyer
P_Ident --> S_Eng
P_Ident --> S_Read

click S_Buyer "/home/start-here/#buyer-operator" "Jump to Buyer path"
click S_Eng "/home/start-here/#engineer-architect" "Jump to Engineer path"
click S_Read "/home/start-here/#reader-analyst" "Jump to Reader path"
click P_HowWorks "/home/how-this-site-works/" "How This Site Works"

🧭 This diagram is a role router: choose your starting identity, then jump directly to the page sequence that minimizes time-to-signal for that role.

If your job is to own risk and outcomes, what sequence of pages turns ambiguity into a safe next step?

flowchart TB
%% Styles (brModel Standard)
classDef i fill:#D3D3D3,stroke-width:0px,color:#000;
classDef s fill:#FFB3B3,stroke-width:0px,color:#000;
classDef p fill:#B3D9FF,stroke-width:0px,color:#000;
classDef r fill:#FFFFB3,stroke-width:0px,color:#000;
classDef o fill:#C1F0C1,stroke-width:0px,color:#000;

S_Buyer("🧑‍💼 Buyer / operator"):::s
I_Problem(["🎯 Decision + constraints + unacceptable failure"]):::i

P_Services("🧰 Services"):::p
P_Case("🧾 Case studies"):::p
R_Risk["🧾 Risk picture + staged roadmap"]:::r

P_Start("📞 Start a conversation"):::p
O_Next("✅ Clear lowest-risk next step"):::o

S_Buyer --> I_Problem --> P_Services --> P_Case --> R_Risk --> P_Start --> O_Next

click P_Services "/services/" "Services"
click P_Case "/case-studies/" "Case studies"
click P_Start "/services/start/" "Start a conversation"

For an operator, the value is not “more information” — it’s risk reduction. You start from the decision you must make, use 🧰 services and 🧾 case studies to surface failure modes, and convert that into a 🧾 risk picture + roadmap before you invest further.

If you build systems, what path gets you from “interesting idea” to a concrete architecture you can implement and govern?

flowchart TB
%% Styles (brModel Standard)
classDef i fill:#D3D3D3,stroke-width:0px,color:#000;
classDef s fill:#FFB3B3,stroke-width:0px,color:#000;
classDef p fill:#B3D9FF,stroke-width:0px,color:#000;
classDef r fill:#FFFFB3,stroke-width:0px,color:#000;
classDef o fill:#C1F0C1,stroke-width:0px,color:#000;

S_Eng("🧑‍💻 Engineer / architect"):::s
I_Stack(["🧩 What must be enforced and why?"]):::i

P_Prim("📐 Methodology"):::p
P_Traces("🧭 CausalGraphRAG"):::p
P_Constr("🔒 Constraints & SHACL"):::p

R_Design["📐 Architecture primitives + enforcement model"]:::r
O_Build("✅ Buildable, auditable system design"):::o

S_Eng --> I_Stack --> P_Prim --> P_Traces --> P_Constr --> R_Design --> O_Build

click P_Prim "/methodology/" "Methodology"
click P_Traces "/methodology/causalgraphrag/" "CausalGraphRAG"
click P_Constr "/methodology/constraints/" "Constraints & SHACL"

For an engineer, the causal pivot is simple: constraints + provenance + traces turn an LLM from “chat” into a system that can be governed. This path gives you the primitives you need to reason about correctness, not just capability.

If you’re learning, what path gives you reusable mental models (not just opinions) as fast as possible?

flowchart TB
%% Styles (brModel Standard)
classDef i fill:#D3D3D3,stroke-width:0px,color:#000;
classDef s fill:#FFB3B3,stroke-width:0px,color:#000;
classDef p fill:#B3D9FF,stroke-width:0px,color:#000;
classDef r fill:#FFFFB3,stroke-width:0px,color:#000;
classDef o fill:#C1F0C1,stroke-width:0px,color:#000;

S_Read("🧑‍🔬 Reader / analyst"):::s
I_Why(["❓ What causes failure and what fixes it?"]):::i

P_ReadHow("📝 How to read"):::p
P_Blog("📰 Blog"):::p
P_Topics("🏷️ Topics"):::p

R_Models["🧾 Causal models + diagrams + counterfactuals"]:::r
O_Transfer("✅ Transferable frameworks"):::o

S_Read --> I_Why --> P_ReadHow --> P_Blog --> P_Topics --> R_Models --> O_Transfer

click P_ReadHow "/blog/how-to-read/" "How to Read Posts"
click P_Blog "/blog/" "Blog"
click P_Topics "/blog/topics/" "Topics"

For analysts, the output is a model you can argue with: explicit mechanisms, counterfactual claims, and diagrams that connect evidence to decisions. That’s the fastest way to build signal without getting trapped in “AI vibes”.