jeremy-dodson commited on
Commit
ec0e589
·
verified ·
1 Parent(s): 1f351d2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +54 -6
README.md CHANGED
@@ -1,10 +1,58 @@
1
  ---
2
- title: README
3
- emoji: 👀
4
- colorFrom: red
5
- colorTo: purple
6
  sdk: static
7
- pinned: false
 
8
  ---
9
 
10
- Edit this `README.md` markdown file to author your organization card.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ title: Piqued AI
3
+ emoji: ⚡
4
+ colorFrom: purple
5
+ colorTo: blue
6
  sdk: static
7
+ pinned: true
8
+ short_description: Governance-aware AI · Open standards · Local-first
9
  ---
10
 
11
+ ![Piqued AI](./pai-logo.png)
12
+
13
+ # Piqued AI
14
+
15
+ Governance-aware AI for collaborative, regulated environments.
16
+
17
+ ## What we build
18
+
19
+ - **Open standards** — [pqdx](https://pqdx.dev) (universal state protocol)
20
+ and [pqdc](https://pqdc.dev) (software-development domain profile).
21
+ - **Two model families** — base substrates (SymbiOS-Core) and curation
22
+ engines (Synapse-Engine), published on Hugging Face for customers to
23
+ fine-tune privately on their own corpus.
24
+ - **Local-first deployment** — designed for on-prem, edge, and air-gapped
25
+ environments. Your data stays on your substrate.
26
+
27
+ ## Models
28
+
29
+ Two parallel families, sized for community fine-tuning and on-prem inference.
30
+
31
+ ### SymbiOS-Core — memory-aware base substrate
32
+
33
+ Foundation models with the scaffolding for memory-awareness baked in.
34
+ Published as **starter substrates** — generic, org-naive, ready to fine-tune.
35
+ Each customer's operational substrate is a private fine-tune produced by
36
+ running Synapse-Engine over their own corpus; that private model stays on
37
+ the customer's substrate and never publishes.
38
+
39
+ Sizes: `1.5B` · `7B` · `14B` · `32B`
40
+
41
+ ### Synapse-Engine — curation and orchestration
42
+
43
+ Instruct-tuned, MCP-native models that take raw corpora (conversations,
44
+ decisions, governance artifacts) and turn them into the curated training
45
+ data that produces a SymbiOS-Core fine-tune. The methodology made
46
+ executable.
47
+
48
+ Sizes: `1.5B` · `7B` · `14B` · `32B`
49
+
50
+ > *All repos currently private — namespace reservation phase. Weights,
51
+ > model cards, and licenses publish per-tier as each model ships.*
52
+
53
+ ## Contact
54
+
55
+ - **Piqued AI** — [piqued-ai.com](https://piqued-ai.com)
56
+ - **Parent (holding)** — [Our Collaborative Space](https://ourcollaborative.space)
57
+ - **Sister entity** — [Piqued Solutions](https://piqued-solutions.com) —
58
+ cybersecurity, DevSecOps, AI advisory