An AI-ready knowledge base that answers from your own records, and shows its sources.
We turn your procedures, policies and records into one checked, permission-controlled source that staff and AI tools such as Microsoft 365 Copilot answer from, with answers that cite the documents they came from, so staff can check them in one click. Answers grounded in your own documents, and we run Jundra on one ourselves.
Opens a prepared email to admin@jundra.com. The first consultation is free; discovery is a paid stage, scoped in writing. How it works
What is an AI-ready knowledge base?
An AI-ready knowledge base is your organisation’s procedures, policies and records turned into one checked, owner-maintained, permission-controlled source that staff and AI tools such as Copilot answer from, with links to the documents behind their answers.
Read this diagram as a list
- Today, know-how is scattered: procedures, policies, reports and records spread across drives, inboxes and a few people’s heads.
- AI organises your documents into an AI-ready knowledge base: clear, linked pages on your topics, sites, assets, people and decisions, each linking to its source.
- Staff and AI agents ask questions in a chat assistant, your intranet or your team chat.
- They get answers from your own material, with a link to each source. Gaps and conflicting documents are flagged, not hidden.
- People approve updates: new documents go into a source folder, AI drafts the updates and a named owner in your team approves them before they go back into the knowledge base. A scheduled health check flags stale pages, contradictions and missing topics.
What makes it AI-ready, how it helps and how you maintain it
Read this diagram as a list
- Clear, linked pages on your topics, sites, assets, people and decisions.
- A source for every page: each page links to the documents it came from, and facts, decisions and recommendations are marked as such.
- Open formats in your own accounts, with full version history.
- Agreed data rules: what is in scope, what is sensitive and what must stay out.
- Named owners approve: AI drafts updates and a named owner in your team approves them.
- A scheduled health check flags stale pages, contradictions and missing topics.
How it helps you
- Answers grounded in your own approved material, with links to the documents behind them.
- Consistent answers from the same approved sources, whoever asks.
- New starters find the approved procedure themselves instead of asking around.
- Conflicting documents and gaps are flagged, not hidden.
- Copilot and other AI tools become more useful, because they answer from checked pages.
- A foundation your AI agents can work from.
How you use and maintain it
Staff ask in Copilot, Teams, your intranet or another AI assistant. AI reads each document once and drafts page updates as new material arrives; your named owner approves them. A scheduled health check flags stale pages, contradictions and missing topics. We train your owners and hand over the runbook.
Organisations whose know-how is spread across drives, inboxes and people.
Government agencies, infrastructure operators, not-for-profits and established businesses with years of procedures, reports and records, who want AI tools that know their business and their rules.
Signs you need it
- Staff keep asking the same questions, and the answer depends on who you ask.
- You have Copilot or another AI assistant, but nobody trusts its answers, or you worry what it can see.
- SharePoint and shared drives hold several versions of the same procedure.
- Key know-how sits with a few long-serving people.
A knowledge base built for people and AI.
The exact deliverables are agreed in your written proposal.
- Source, risk and permission map
Which documents and systems are in scope, what is sensitive and must stay out, and who can see what today, including what is over-shared. - Clean, linked pages
Plain-language pages on your topics, sites, assets and decisions, in open formats you keep, with full version history. - Answers that cite their sources
Staff and AI assistants answer from those pages, with links to the documents behind them. Where the sources don’t cover a question, it is designed to say so rather than guess. - Permission-aware, in the tools you use
Retrieval is built on your existing access groups and tested before go-live, so the knowledge base does not widen who can see what, through Copilot, Teams, your intranet or your own AI assistant. - Evaluated before go-live
Tested against a set of real questions from your staff, with the results shown to you before anyone relies on it. - Kept current by your team
AI drafts updates as documents change; your named owner approves them. Runbook, health checks and training handed over.
Six steps, from one team to the whole organisation.
We start with one team or topic, test it against the questions your staff really ask, then scale what works.
STEP 01
Free consultation
30 minutes, free30 minutes on your knowledge problem and whether an AI-ready knowledge base is the right fit. No charge, no obligation.
STEP 02
Discovery (paid)
paid, scoped in writingWe sample your sources, systems, permissions and data rules with your team and agree what is in and out of scope. A paid stage, scoped and priced in writing before it starts.
STEP 03
Written proposal
scope, steps, priceScope, steps, timeframe, deliverables and price, in writing.
STEP 04
Pilot or MVP
fixed deliverablesWe build the knowledge base for one team or topic, test it against real questions and show you the results.
STEP 05
Rollout
scale what worksWe extend it to more teams and sources, and connect it to the tools you use.
STEP 06
Handover and support
your team, your filesYour files in your accounts, training for the people who maintain it, 60-day defect fixes and optional support.
Jundra runs on its own LLM wiki.
Our project decisions, sources and lessons go into our own LLM wiki, and our AI agents read it before they start work. Work starts from what we already know, decisions trace back to their source, and specialists joining a project get up to speed quickly.
Changes are tracked in version history, and a person reviews what the agents write. That is the discipline we bring to yours. We do not publish what is in our wiki.
Want to test it on one team’s documents? Book a free 30-minute consultation →
What you can count on.
Free first conversation
A 30-minute consultation costs nothing and commits you to nothing. Discovery is a paid stage, scoped and priced in writing, and the pilot or MVP has fixed deliverables.
60-day defect fixes
We fix defects in our work for 60 days after delivery, at no charge, on top of your rights under the Australian Consumer Law.
People stay in charge
AI drafts and people approve. Nothing is added to your knowledge base without a named owner reviewing it.
Prequalified and insured
A prequalified supplier under the NSW Government ICT Services Scheme (SCM0020), with professional indemnity, public and products liability cover.
Built to fit Australian AI guidance.
We design to the National AI Centre’s Guidance for AI Adoption, and prepare the paperwork your own review needs: an AI register entry, a risk and impact summary, test results and the points where a person approves. That gives your team ready-made evidence for:
- the NSW AI Assessment Framework (NSW agencies);
- the Australian Government’s policy for the responsible use of AI in government (Commonwealth agencies);
- the National AI Centre’s Guidance for AI Adoption (any organisation);
- the OAIC’s privacy guidance on commercially available AI products.
Your privacy, legal and risk owners make the decisions; we give them the information to make them quickly.
Ownership of the work, data privacy and pricing are answered in the questions below.
Questions we get asked.
We already have Copilot. Why do we need this?
Copilot answers from anything each user can already open in your Microsoft 365, including old drafts, duplicates and files shared too widely. We give it, and the other AI tools you use, a checked, current, permission-controlled source to answer from, and fix the over-sharing before more people start asking questions.
Will staff see documents they shouldn’t?
No more than they can today. Retrieval is built on your existing access groups, and we test it with real user accounts before go-live. Before go-live we review who can see what and flag files that are shared too widely, so the knowledge base does not expose them.
How does it work? RAG, LLM wiki and knowledge graph explained.
RAG (retrieval-augmented generation) is the core technique: the AI retrieves the right pages from your knowledge base and answers from them, rather than from memory, with a link to each source.
Read this diagram as a list
- A staff member or an AI agent asks a question.
- Search retrieves the relevant pages from your AI-ready knowledge base.
- AI answers from the retrieved pages, not from memory, with a link to each source.
The other terms you may hear: we build the knowledge base as an LLM wiki, where AI maintains linked pages from your sources so knowledge is compiled once and kept current, and add a knowledge graph where relationships between assets, sites, suppliers and rules matter. Platform services call the same thing a knowledge base. We combine them to suit your organisation.
What if the AI gets it wrong?
Answers cite the documents they came from, so staff can check them in one click. Before go-live we test it against a set of real questions and show you the results; where the sources don’t cover a question, it is designed to say so rather than guess. Staff can flag wrong or missing answers, and we fix them at the source page.
Do we have to move our documents or replace systems?
No. We work from the documents and systems you already have, such as SharePoint, Google Workspace, your intranet and existing databases, and agree with you which ones are in scope. The knowledge base sits alongside them.
How do we keep it up to date?
New material goes into a source folder, AI drafts the updates, and your owner approves them. A scheduled health check flags stale pages, contradictions and gaps. We train your team to run it and can provide ongoing support.
More questions (6)
What if our documents disagree with each other?
That is common, and finding it is valuable. Conflicts are flagged for a person to decide, and the decision is recorded with its source.
What about data privacy and confidentiality?
We never put your data into free or consumer AI tools. Before we start, we agree in writing where your data is stored, where AI processing happens, which services may process it and who on our team can see it, to support your obligations under the Privacy Act and the Australian Privacy Principles. Where data needs to stay in Australia, we can use Australian cloud regions where the provider offers them, or run open-source models on your own servers. Where you want a commercial provider, we use business services under terms that do not allow your data to be used for training. We are happy to work under your confidentiality agreement, and the proposal says what happens to copies of your data at the end of the engagement.
You’re a small firm. Can you handle an organisation our size?
Yes. You work directly with Jundra’s project leads, Ken Taggart and Ali Khorsandian, and we bring in specialist developers, engineers and security testers as each project needs, named in your proposal. We start with one team or one task, so the first stage is sized to fit, and we have supported one critical rail monitoring system for more than a decade.
How can we buy from you?
Directly, or through the NSW Government ICT Services Scheme, under which Jundra is a prequalified supplier (SCM0020). Tell us how your organisation buys and we will confirm in writing what applies before discovery starts.
What does it cost?
The 30-minute consultation is free. Discovery is a paid stage, scoped and priced in writing before it starts. After that, scope and price depend on your organisation, so we do not publish package prices: you get a written proposal with the scope, steps, timeframe and price before the pilot starts, and the pilot or MVP has fixed deliverables.
Who owns the work?
You do. You own the project work you pay for, handed over into accounts you own. Pre-existing tools and third-party components, including commercial AI models, stay under their own licences, listed in your proposal. Every project includes a 60-day defect-fix period after delivery, in addition to your rights under the Australian Consumer Law.
Related AI services
Book a free 30-minute consultation.
Tell us about the knowledge your organisation relies on. We will show you what an AI-ready knowledge base could do with it.
You’ll talk with Ken Taggart or Ali Khorsandian, who lead Jundra’s projects. In 30 minutes we will:
- look at one knowledge problem you choose, such as a procedure set, a policy library or one team’s shared drive;
- tell you whether an AI-ready knowledge base, a Copilot set-up or something simpler is the right answer;
- name the main risks to check first: data, permissions and privacy;
- outline what a paid discovery would cover.
No charge and no obligation. If you go ahead, discovery is a paid stage, scoped and priced in writing. Please do not include confidential information in your email.
How this button works
It opens a prepared email to admin@jundra.com in your email app; nothing is booked or sent automatically. Press Send and we reply within one business day to agree a time: a video call, a phone call or a visit to your site. Using webmail? Copy the address into a new email. How we use enquiry details.