Opportunities › Enterprise AI workspace and chat platform › Proposed app
Tillmark
One workspace for every AI model, with team access, spend limits and usage records
- Score
- 58
- confidence 40
- Evidence
- 8
- items on the page
- Rivals
- 4
- named in sources
- First version
- 45 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2023
- Emerging market · Verified
ScreensConcept
Desktop app: Team members chat with AI, but operations and finance teams analyze spend, permissions, and usage records in detail dashboards at a desk. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
Team chat
Routed by rule- TeamFinance Ops
About this appAI proposal
Tillmark is a shared AI chat workspace for teams that use more than one language model. Staff pick a model or let rules choose one. Operations, finance and IT see who used what, what it cost and which data went where.
Admins set which teams may use which models, cap monthly spend per team, and attach shared document libraries and tools. Chats accept text, files and images. Each workspace can run in our cloud or on your own servers if data must stay in house.
It is different from single-vendor team plans because no model provider owns your history, and different from developer frameworks because the admin screens, sign-in and billing views are already built.
- Tagline
- Every model, one set of house rules.
- Voice
- CalmPrecisePlainspoken
- Why this name
- Tillmark joins tilling, steady work on shared ground, with a mark, a record. It suggests a team workspace that keeps a clear record of use.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Model routing rules
Send each request to a model by team, task type or cost ceiling. Switch providers without retraining staff or moving chat history.
No member yet
Team access and permissions
Roles for members, team leads and admins. Control who can use each model, tool and document library, with single sign-on.
No member yet
Spend caps and usage ledger
Set monthly limits per team. Every request is logged by user, model and cost, and exports as CSV for finance.
No member yet
Shared document libraries
Attach company files for retrieval-based answers. Access follows team permissions, and answers cite their source files.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#1F3D36Deep Moss · primary#F6F3ECLinen · surface#1B2421Ink · text#E0A032Amber · accent#B5543AClay · alert#7FA391Sage · data
Typography
Aa
Tillmark
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
PricingAI proposal
The material gives no competitor prices. These tiers are a starting point to test. Model provider fees are billed by each provider or passed through at cost.
Team
$15 a month
- Per user, cloud hosted
- Multi-model routing
- Usage ledger and CSV export
- Shared document libraries
Business
$30 a month
- Per user, cloud hosted
- Spend caps per team
- Single sign-on and roles
- Tool connections
- Audit log
Self-hosted
Contact us
- Run on your own servers
- Annual licence by team size
- Priority support
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
What products in this market make
- TypingMindEstimate$68,108 a month, worth about $1,322,000 at that multiple · getlatka.com
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- Users invite others. A team lead invites colleagues and other teams into shared threads and document libraries. Each new team asks for its own access rules and spend cap, which pulls in the admins.
- Do they become customers?
- Invited colleagues are in the same company, so they can become seats or new team budgets. Test whether invited users trigger a plan upgrade or whether one admin buys everything.
- First customers
- IT or operations leads at companies of 100 to 500 staff who already pay for two or more AI subscriptions and have no single view of cost.
- Where to reach them
- Open-source communities: Join discussions around self-hosted AI workspaces and offer a hosted, supported option to teams who do not want to run it.
- LinkedIn outreach: Message IT and finance leads with a one-page sheet showing how to compare spend across AI subscriptions.
- Founder-led demos: Offer a 30-minute setup where we connect two models and show a real usage ledger.
- First milestone
- Five paying companies of 100 to 500 staff, each with at least two teams using it for 60 days.
Why people already pay for this
- VerifiedGitHub repository gained 1530 stars in 143 days
- Public claimThe repository describes itself as an enterprise AI workspace that supports model routing, meaning organizations can switch between or use multiple AI models.
- Public claimThe platform includes built-in billing and identity management, indicating enterprises need centralized control over AI usage and access.
- Public claimSupport for RAG and MCP (Model Context Protocol) suggests the platform integrates with external data sources and tools, addressing a need for richer AI workflows.
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Model gatewayProvider-neutral LLM routing layerRoutes requests to several model providers and records tokens and cost per call.
- This app · Document searchPostgres vector search with file parsingSupports retrieval over shared files using the existing database.
RisksAI inference
- CommercialOpen-source projects already offer a self-hosted enterprise workspace; buyers may use the free version. Paid value must come from support, governance views and ease of setup.
- PlatformModel providers change APIs, prices and terms often. Routing and cost tracking need constant upkeep.
- DataStoring prompts and company files makes the product a target and raises data handling duties. Customers may demand audits before buying.
- TechnicalSupporting both cloud and self-hosted versions doubles testing and support work for a small team.
- RegulatoryCustomers in regulated sectors may require certifications and data residency terms before they can deploy it.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.