Proposal Generator
Turns a brief plus the knowledge base into a quotation, scope, timeline and bill of materials. Pricing is deterministic — integer paise, GST-aware, no model in the arithmetic.
From knowledge to proposals, automation and deployment.
WorkKit turns how a team actually delivers — rates, specifications, processes, assumptions, exclusions — into quotations, proposals and bills of quantity. The rules do the arithmetic. The model only writes the prose.
What is WorkKit
Anyone can call a language model. The hard part is having something true to hand it.
Every experience technology project produces the same artefacts — a bill of quantities, a cost estimate, a quotation, an invoice. The knowledge inside them is the most valuable thing a delivery team owns, and it is almost always scattered across old spreadsheets and the memory of whoever quoted last time.
WorkKit is the place that knowledge lives: rates, equipment specifications, installation processes, standing assumptions and exclusions, each one a versioned entry that records where it came from and when it was last verified. Proposal generation is the first thing built on top of it — one application of the knowledge base, not the product.
Core capabilities
Each is a module under the same contract — it deploys, versions and rolls back on its own, and inherits identity, logging, metrics and backup from the platform.
A brief becomes a quotation, a scope document, a timeline and a bill of materials. Line items, discounts and GST — including the CGST/SGST versus IGST split — are computed by a tested engine in integer paise, never by the model.
Rates, hardware specifications, processes and legal standing terms as structured entries. Each carries provenance and a verification date, and anything still unverified is refused rather than quietly served.
Bills of quantity and equipment selection drawn from the same entries that priced the job, so the specification a client signs and the list procurement orders cannot drift apart.
Solutions
WorkKit is written by a team that installs this equipment. Every domain below is a body of knowledge the platform holds — specifications, rates, processes and the assumptions that go wrong when nobody writes them down.
Sensor-driven and touch-driven experiences where the hardware, the content and the failure modes all have to be specified before anyone quotes a price.
Large-format projection across irregular surfaces, where lumens, throw distance, blending and rigging determine both the design and the budget.
Identity and interaction at scale — visitor journeys, access, personalisation and the reader network that has to stay up for the duration of an event.
Connected devices, show control and the network underneath them — the part that is invisible when it works and the entire project when it does not.
Permanent installations judged over years rather than days, where maintainability, warranty and spares matter as much as the opening night.
How WorkKit works
A language model asked to total a quotation will occasionally be wrong, will be wrong plausibly, and the error ends up in a document someone signs. So it is never asked.
What it does
Holds rates, equipment specifications, installation processes, assumptions and exclusions as versioned entries, each recording its source and verification date.
What it never does
Never infers. A document with no link to a project stays unfiled rather than being filed under the nearest plausible one.
Labs
Each entry says exactly where it stands — including when that is “has never run”. Nothing here is claimed to be finished.
Turns a brief plus the knowledge base into a quotation, scope, timeline and bill of materials. Pricing is deterministic — integer paise, GST-aware, no model in the arithmetic.
Versioned operational knowledge with provenance on every figure — retrieved by exact id rather than similarity, so “why did it quote that rate?” has a one-line answer.
Letting a model write proposal prose around figures it was handed, and checking afterwards that it invented neither a number nor a piece of equipment.
The platform is not generally available yet. If you run delivery for installations, events or permanent experiences, we want the proposals you already send — they are what the knowledge base is derived from, and early teams shape what gets built next.