Project Knowledge and Lessons Learned Agent
The Project Knowledge and Lessons Learned Agent builds a searchable knowledge base from your completed projects: priced BoQs and rates, programmes and actual durations, claims and their outcomes, subcontractor performance and close-out lessons. Your team asks questions in plain language and gets answers with citations to the source documents, so estimates, tenders and programmes draw on what actually happened rather than on memory.
Ideal forEstimators
The difference
From 1-2 hours per question to 5-10 minutes
Estimators and planners find the relevant precedent in minutes during short tender periods instead of digging through old project folders.
Done by hand
1-2 hours per question
With the Vanine agent
5-10 minutes
Time returned to your team
90%
Indicative figures. We measure your own baseline in the first fortnight so the numbers you see are yours.
Experience that stays when people leave
Hard-won knowledge often walks out of the door with a senior QS or project manager. The agent keeps it in a form the whole team can use.
Sharper estimates and programmes
Real rates and actual durations from similar projects make tenders more competitive and programmes more realistic. Every figure is traced back to its source project.
The same mistakes not repeated
Lessons from disputed claims, failed subcontractors and design issues are surfaced when a similar project starts, not buried in a close-out report no one reads.
Capabilities
Every past project working for your next estimate, tender and programme.
Built for estimators, quantity surveyors, planners, project managers and directors at South African contractors, project management firms and consulting practices with years of project records.
Project archive indexing
Reads past BoQs, cost reports, programmes, minutes, claims and close-out reports from SharePoint, Procore, Aconex and estimating systems and indexes them by project type, region and year.
Rate and cost benchmarks
Finds rates and elemental costs from comparable past projects and shows the source, date and contract context so estimators can adjust them for current conditions.
Programme benchmarks
Compares planned and actual durations for activities and project types to inform realistic programmes and extension of time risk.
Claims and dispute history
Summarises past variations, extension of time claims and compensation events and how they were resolved, by contract form and client.
Subcontractor performance
Collates records on subcontractor quality, programme and safety performance across projects to support selection and pre-qualification.
Cited answers
Answers questions in plain language with links to the source documents, so every figure and lesson can be checked before it is relied on.
How it works
From your systems to a result you sign off
Access follows your existing SharePoint and project permissions. Your estimators and planners decide how benchmarks are used, and pricing decisions stay with the bid team.
Connect the places your project archives already live.
Project archives in SharePoint, Procore and Aconex, estimating data from Candy, WinQS or BuildSmart, and programmes from Primavera P6 or MS Project. Already use something else? We connect that too.
- OpenAI Frontier
- Your own AI platform or models
- Microsoft 365 (Teams, SharePoint, Outlook, Excel)
- Copilot Studio
- Candy, WinQS and BuildSmart
- Primavera P6 and MS Project
- Procore and Aconex archives
Answers questions on request in Teams, Copilot Studio or OpenAI Frontier, and updates the knowledge base as projects close out.
Cited answers, rate and duration benchmark tables in Excel, and lessons learned summaries for tender and project start-up meetings.
Step by step
- 01
The agent indexes your completed project records and updates as new projects close out.
- 02
A team member asks a question in Teams or Frontier while estimating, tendering or planning.
- 03
The agent searches rates, programmes, claims and lessons from comparable projects.
- 04
It returns a cited answer or benchmark table with links to the source documents.
- 05
Your estimator, planner or project manager reviews and applies the findings.
Deployment
Runs inside the tools your team already uses
The same agent and the same cited output, delivered three ways: through the AI assistant you have rolled out, on a model you host yourself, or inside your own software.
In the frontier assistant you already pay for
We publish the agent as an MCP server. Add it as a connector and your team calls it from the chat they use every day, with no new tool to learn.
"Search past projects for earthworks rates"
- question
- Earthworks rates
- archive
- 120 projects
On a model you host yourself
Run the agent on open-weight models in your data centre or private cloud. Your project archives never leave your network, which keeps POPIA and data residency straightforward.
- Llama
- Mistral
- Ollama
- or any OpenAI-compatible endpoint
Inside your own software
Trigger a run from your own workflows, and push the finished output into your ERP, GRC or reporting systems. We build the integration with your team.
- REST API
- Webhooks
- Batch jobs
- Embeddable review panel
Governed the same way, every route
Whichever way your team reaches the agent, identity, permissions and audit stay with your existing controls.
- Single sign-on through Entra ID, Okta or Google Workspace
- Reads only the documents each user is already allowed to open
- Every tool call written to an audit log you can export
How we deliver it
One agent, three ways to hold it
Start where your team already works, then take it as far into your systems as the value justifies.
DATA SECURITY & PRIVACY
Your project archives stay yours.
The agent runs inside your tenancy, reads only what each task needs, and never trains on your project archives. Every step is logged so your auditors can see exactly what it touched.
Built for regulated work.
+YOUR PROJECT ARCHIVES NEVER LEAVE YOUR ENVIRONMENT
More agents
More agents we build for South African enterprises
Each one is built on the same foundation: your documents, your rules, and your people signing off.
Precision AI for Institutional Workflows
