ProjectVIEW AI: AI Agents and Assistant for ERP

AI Assistant

How AI Works on ERP Data

AI Agents

MCP Server

AI API

Models: SLM + LLM

Data Security & Governance

Deployment: Cloud or On-premises

Why ProjectVIEW AI

AI Roadmap

What is ProjectVIEW AI?

DANAOS ProjectVIEW AI is the intelligence layer for project-based companies that run on ProjectVIEW ERP. ProjectVIEW AI is the AI add-on to ProjectVIEW ERP: agents and an assistant that work on ERP data, combining domain-specific small language models with leading LLMs. It is built for construction, shipbuilding and repair, marine and offshore, mining and quarrying, and project-based manufacturing.

AI that knows your business.

Generic AI tools know language. They do not know your budget, your commitments or your subcontractor certifications. ProjectVIEW AI reads the same BoQ, WBS and cost code structure your teams already work in. That is why its answers refer to your projects, your rates and your approvals, not to averages from the internet.

ERP records what happened. AI explains why it happened and what will happen next. ProjectVIEW AI is designed to detect risks early, explain them in plain language and prepare the next step. A person approves every consequential action.

How does ProjectVIEW AI relate to ProjectVIEW ERP and ProjectVIEW OS?

ProjectVIEW ERP is the integrated ERP built on the BoQ ↔ WBS ↔ Cost Codes framework. It is the system of record. ProjectVIEW AI is the intelligence layer on top of it. Together they are positioned as ProjectVIEW OS. ProjectVIEW OS is a positioning name, not a separate product.

Layer	Role	What it does
ProjectVIEW ERP	System of record: process and data layer	Runs estimating, procurement, cost control, subcontractors, materials, machinery, payroll and finance on one data set. Deterministic: transactions, approvals, audit trail.
ProjectVIEW AI	Intelligence layer	Answers questions, detects anomalies, forecasts, explains and recommends on ERP data. Probabilistic: every consequential action goes to a person for approval.
ProjectVIEW OS	ERP + AI together	The combined platform: system of record plus system of intelligence, under one governance model.

Deterministic ERP plus probabilistic AI is the point. The ERP keeps the numbers right and controlled. The AI makes them readable and forward-looking. For the reasoning behind this design, read why we built it this way and why AI agents fail without ERP. For comparisons, see comparison with generic AI copilots and comparison with AI point tools.

What can you do with ProjectVIEW AI?

There are four ways to work with ProjectVIEW AI. Each one uses the same ERP data and the same role permissions.

Ask the assistant. Ask questions of your ERP data in plain language, for example “Which projects are most likely to exceed budget?” or “Show me my pending approvals.” See the AI Assistant.
Run agents. Purpose-specific agents, being rolled out by module, are designed to watch a process, such as procurement or cost control, flag what is off, explain it and request approval before anything is executed. See AI Agents.
Connect Claude or ChatGPT through MCP. MCP is an open standard for connecting AI assistants to business systems. The ProjectVIEW MCP connector lets users ask questions of their own data, build dashboards and KPIs, and build their own agents. See the ProjectVIEW MCP server and the AI API.
Work by voice and headless. Run operations without opening ERP screens. ProjectVIEW AI is designed to let a site engineer dictate the daily log, transcribe it, link it to BoQ, WBS and cost codes, and prepare the ERP entries for review; availability is confirmed per deployment. People speak. AI interprets. ProjectVIEW ERP structures, validates and controls. See how it works on ERP data.

Which ERP modules do the AI agents work on?

ProjectVIEW AI works across the ProjectVIEW ERP modules. Each agent is designed for one process and reads the data of that module plus the related cost structure.

ERP area	What ProjectVIEW AI is designed to do	Agent
Budget, estimation and bidding	Find comparable items and Cost Recipes, flag unusual assumptions	Bidding & Estimation
Cost control	Forecast cost at completion, margin erosion, probability of overrun, EVM	Cost Control & Forecast
Subcontractors	Track delayed certifications, retention exposure, commercial risk	Subcontractor & Certification
Procurement and purchasing	Compare quotations, detect anomalies, check against budget rates	Procurement
Materials and production	Follow material flow and production progress against the WBS	Materials & Production
Machinery	Flag abnormal fuel use, idle time and CO₂ per activity	Machinery & Maintenance
Accounting and finance	Best, expected and worst-case cash flow forecasts	Cash Flow & Finance
Contracts, claims and variations	Surface unapproved variations and assemble evidence for claims	Claims & Variations

Who is it built for?

ProjectVIEW AI is for the people who carry cost, schedule and commercial risk on projects:

C-level executives and directors who need a short, reliable briefing across all projects.
Project managers, cost controllers and quantity surveyors who need early warnings, not month-end surprises.
Estimators and bidding managers who want comparable items and rates from past work.
Procurement officers, subcontractor managers and contracts managers who check quotations, certifications and variations.
Plant and machinery managers, production managers and finance officers.

Industries: construction, shipbuilding and ship repair, marine and offshore, mining and quarrying, and project-based manufacturing. ProjectVIEW ERP clients include VINCI (construction) and Papilas Shipyard (shipbuilding).

How is ProjectVIEW AI governed?

AI should operate inside enterprise governance, not around it. Every request follows one chain: Identity → Role → Permission → Data → AI → Action. ProjectVIEW AI sees only what the user’s role is allowed to see, and it cannot approve what the user cannot approve.

Data is not sent to public consumer AI services. ProjectVIEW AI runs in a private environment.
On Microsoft Azure, prompts are not used for model training; data residency options, encryption at rest and customer-managed keys are available.
ERP context is retrieved at inference time, so models do not need to be retrained on client data.
Consequential actions always require human approval, and every step is logged.
DANAOS Projects holds ISO 27001, ISO 9001, ISO 14001 and ISO 20000-1.

Read more on security and governance, the models we use and AI risk management.

How is it licensed and bought?

There are three ways to buy:

ProjectVIEW ERP only. The ERP, with an implementation time of 5 to 7 months.
ProjectVIEW AI on top of your existing systems. Connect SAP, Oracle, Primavera P6, MS Project, BIM models and documents; data is mapped to BoQ, WBS and cost codes.
Both, as ProjectVIEW OS. ERP and AI on one data set and one governance model.

Pricing depends on the modules, the number of users and the deployment option. AI usage can be metered with DANAOS AI Credits. Deployment runs on Microsoft Azure (SaaS) or on-premises; see deployment options. For what comes next, see the AI roadmap.

Frequently asked questions

Is ProjectVIEW AI a separate product from ProjectVIEW ERP?

ProjectVIEW AI is an add-on with its own commercial model. It works best on ProjectVIEW ERP data, and it can also run on top of existing systems such as SAP, Oracle or Primavera P6. ProjectVIEW OS is the name for ERP and AI together.

Is our data used to train AI models?

No retraining on client data is required. ProjectVIEW AI retrieves the relevant ERP context at inference time. On Microsoft Azure, prompts are not used for model training.

Can ProjectVIEW AI agents act on their own?

Agents detect, investigate, explain and recommend. Before any consequential action they request approval from the responsible person. Every step is recorded for audit.

Is ProjectVIEW AI available now?

Partly. ProjectVIEW AI in daily processes, including natural-language questions on ERP data, and the MCP connector for Claude and ChatGPT are available now with ProjectVIEW ERP v12, released in October 2026 for all clients. The purpose-specific agents are being rolled out by module and configured per client. Ask us in the demo which agents apply to your modules.

Which AI models does ProjectVIEW AI use?

It is model-flexible. Domain-specific small language models handle pattern recognition and numerical analysis; leading LLMs handle reasoning, language and multi-step planning. The model is chosen per use case.

Want to see it on your own data?

See ProjectVIEW AI work on project data that looks like yours: budgets, commitments, certifications and forecasts. Book a demo with our team.

Book a Demo