From 2d3391f531ab593928cdade762103bf7642cd165 Mon Sep 17 00:00:00 2001 From: Jonathan Date: Sun, 19 Apr 2026 20:31:51 +0200 Subject: [PATCH] feat: product page with architecture and platform framing --- src/pages/product.astro | 424 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 424 insertions(+) create mode 100644 src/pages/product.astro diff --git a/src/pages/product.astro b/src/pages/product.astro new file mode 100644 index 0000000..c0e94c2 --- /dev/null +++ b/src/pages/product.astro @@ -0,0 +1,424 @@ +--- +import AppLayout from '../layouts/AppLayout.astro'; + +const user = Astro.locals.user; + +const tiers = [ + { + name: 'Fenja Basic', + desc: 'General knowledge work — document retrieval, Q&A, summarisation, and search across the organisation\'s corpus. The foundation every other tier is built on.', + current: true, + }, + { + name: 'Fenja Code', + desc: 'Developer-assisted coding within the customer\'s own codebase. The agent understands project structure, conventions, and prior decisions.', + current: false, + }, + { + name: 'Fenja Analyze', + desc: 'Strategic data analysis — structured extraction across large document corpora, with full provenance on every finding.', + current: false, + }, + { + name: 'Fenja Agentic', + desc: 'Governed agent-based automation. Non-technical users define repeatable workflows; the platform handles execution, scheduling, and audit.', + current: false, + }, +]; +--- + +
+ + +
+

Product

+

+ The platform for sovereign AI. +

+
+ + +
+
+

+ Fenja is not a single product. It is a platform for running advanced AI + inside the customer's own infrastructure. It solves many use cases because + it provides the building blocks every regulated AI deployment needs: a local + model, agents that do work, a layer that teaches the AI the organisation's + language, and the tools those agents use to act. +

+

+ One deployment, fully under your control, with no data ever touching + external infrastructure. +

+
+
+ + +
+

How it is structured

+ +
+ +
+
+ AI Model +
+
+

+ Locally hosted, open-source foundation model (Llama 3, Mistral). + No data ever leaves the customer's infrastructure. +

+
+
+ Foundation +
+
+ +
+
+ Agents +
+
+

+ Specialised AI agents that handle distinct tasks — retrieval, summarisation, + analysis, drafting — and collaborate to solve complex queries. +

+
+
+ Execution +
+
+ +
+
+ Semantic Layer + Fenja Knowledge +
+
+

+ The business vocabulary through which the AI understands the organisation — + its structure, terminology, decisions, and the people behind them. + Grows automatically as the organisation uses the platform. +

+
+
+ Context +
+
+ +
+
+ Tools +
+
+

+ The capabilities agents use to act: read files, query systems, search the corpus, + produce structured output, trigger processes. +

+
+
+ Capabilities +
+
+ +
+
+ + +
+

Why this is a platform, not a product

+
+

+ The same sovereign infrastructure supports fundamentally different shapes of work. + One deployment, one security model, one audit trail — but the use cases it enables + expand as you add tiers. +

+
+
+ {tiers.map((tier) => ( +
+
+ {tier.name} + {tier.current && ( + Bifrost MVP + )} +
+

{tier.desc}

+
+ ))} +
+
+ + +
+

What Project Bifrost delivers first

+
+

+ The MVP in Project Bifrost is Fenja Basic: a locally hosted LLM with agent + flows, plus Fenja Knowledge — the semantic layer that lets the AI understand + not just your documents, but the context behind them. +

+

+ This is the foundation. It solves real work today. And every tier that follows — + Code, Analyze, Agentic — is built on it without requiring a new deployment, + new infrastructure, or new compliance review. +

+
+
+ + +
+

The platform extends to

+
    +
  • + Fenja Code + + Developer-assisted coding inside your own infrastructure — source code never leaves. + +
    • +
  • + Fenja Analyze + + Structured data extraction and analysis across large corpora, with provenance intact. + +
    • +
  • + Fenja Agentic + + Governed automation for repeatable workflows — defined in natural language, audited end to end. + +
    • +
+
+ +
+ + +