AI Proposal Engine
Customer context, design scope, pricing logic, scope of work, contract terms, and signature flow become structured proposal memory.
Landscape Construction OS
The AI operating system for landscape construction.
Landscape OS is Hardac's first vertical product, built to prove a specific operating thesis: one owner can run a $10M landscape construction company with AI handling the operational layer around proposals, schedules, field work, receipts, job costing, and owner briefing.
That is the target, not a claim that the model is fully achieved. AI does not replace field crews, subcontractors, or owner judgment. It takes on more of the operating work around them.
The market problem
Why landscape construction breaks
High-ticket outdoor-living and landscape construction companies do not fail because they lack another dashboard. The hard work lives between systems: customer context, design scope, estimates, proposals, schedules, crews, subcontractors, materials, weather, field updates, receipts, job costing, and owner judgment.
Traditional software records pieces of that work. Landscape OS is being built to run more of the operating layer.
First wedge
The AI Proposal Engine
The AI Proposal Engine turns customer context, design scope, pricing logic, scope of work, contract terms, and signature flow into signable proposals for large outdoor-living projects. The proposal cycle has moved from weeks to hours.
This matters because the proposal is not just a sales document. It becomes the structured memory of what was sold, what needs to be built, what the customer expects, and what operations must deliver.
Expansion path
From proposal to operations
Landscape OS starts with the commercial wedge and expands into the operating system around it: proposal memory, sold pipeline, project operations, field truth, job costing, and owner briefing.
Contract and scope memory
The signed proposal becomes the operating record of what was sold, what the customer expects, and what must be delivered.
Sold pipeline
Upcoming work moves from sales output into a live pipeline that can be broken into scopes, crews, materials, and timing constraints.
AI Project Manager
Projects are sequenced, coordinated, and re-sequenced as labor, weather, suppliers, and field conditions change.
Field and finance loop
Field updates, receipts, job costing, and owner briefing pull operational truth back into the system.
Product surface
What Landscape OS owns
Landscape OS is being built around the operating loop of a landscape construction company: from first customer context through owner briefing after work is underway.
01In flight
Lead
02Planned
Design
03Live
Estimate
04Live
Proposal
05Live
Contract
06In flight
Schedule
07In flight
Crew
08In flight
Field Updates
09Live
Receipts
10In flight
Job Costing
11Planned
Owner Brief
AI operating layer
Owner
- Proposals
- Customer Communications
- Project Management
- Scheduling
- Materials
- Receipts
- Job Costing
- Finance
- Owner Briefings
Owner
- Proposals
- Customer Communications
- Project Management
- Scheduling
- Materials
- Receipts
- Job Costing
- Finance
- Owner Briefings
Next module
The AI Project Manager
The AI Project Manager takes the sold pipeline, breaks projects into scopes, learns crew and subcontractor production rates, sequences work, coordinates material timing, and re-sequences when weather, labor, or supplier delays change the plan.
This is in flight and planned, not fully shipped. The direction is clear: move from project tracking to AI-supported operations while keeping owner judgment in the decisions that still require it.
Operating modules
Built system by system, against live work
Each module is a path from recorded information to owned operating work. Shipped modules are marked live; planned and in-flight modules are labeled accordingly.
First wedge
LiveAI Proposal Engine
What it owns
Customer context, design scope, pricing logic, scope of work, contract terms, and signature flow.
Why it matters
The proposal is not just a sales document. It becomes structured memory for what was sold, what needs to be built, what the customer expects, and what operations must deliver.
Expansion module
In flight / plannedAI Project Manager
What it owns
Sold pipeline intake, scope breakdown, crew and subcontractor production-rate learning, sequence planning, material timing, delay response, re-sequencing, and escalation where owner judgment is required.
Why it matters
This is the path from proposal automation to AI-run operations. The module is in flight and planned, not fully shipped.
Customer context
In flightCustomer Communications
What it owns
Follow-ups, updates, reminders, conversation history, context-aware replies, and commitments tied to the job.
Why it matters
Customer context stays attached to the work instead of living in the owner's memory.
Operations
In flight / plannedScheduling and Materials
What it owns
What was sold, what needs to be built, who can build it, which materials are needed, and when delays require a schedule change.
Why it matters
The schedule becomes a living operating plan, not a static calendar.
Field to finance
Receipts liveField Updates and Receipts
What it owns
Receipt photos, job notes, project updates, materials, expenses, extraction, project matching, and job financial entry.
Why it matters
Field inputs flow back into the operating system. Receipt reconciliation is live, so expenses can be identified and tied back to the right job.
Financial truth
In flightJob Costing and Finance Ops
What it owns
Job costing, expense categorization, reconciliation, margin visibility, and owner reporting.
Why it matters
Financial truth stays closer to live field activity. This is an in-flight module, not a claim of fully automated finance operations.
Owner interface
PlannedOwner Briefing
What it owns
What needs attention, what changed, what is blocked, and which decisions still require owner judgment.
Why it matters
The owner gets an operating interface, not another dashboard to manage.
Live operating lab
Why Gillespie matters
Gillespie is not a demo account. It is the live operating lab where Landscape OS is tested against real customers, real jobs, real field inputs, real owner decisions, and real operational constraints.
Every workflow shipped removes work from the company today and teaches Hardac what an AI-run landscape construction company needs to become tomorrow.
Company thesis
Why this can become a company
The first goal is to prove the operating model inside Gillespie Landscape. The next goal is to repeat the modules across similar high-ticket landscape construction and outdoor-living companies.
The opportunity is not a one-off internal tool. It is a repeatable vertical operating system built from live operational proof.
Build engine
Built by Forge
Landscape OS is built by Forge, Hardac's AI-native build engine. Forge coordinates product, architecture, engineering, QA, documentation, and delivery agents so Hardac can build and iterate with speed and rigor.
Forge is the engine behind the work. Landscape OS is the first commercial wedge.
Design partners
Help prove the Landscape Construction OS
Hardac is looking for operators, design partners, and investors who understand the operational weight of landscape construction and want to help prove where AI operators should own the work.