AI Manager — run your week with AI
Every business gets its own workspace: the AI reads the situation, proposes the week's plan and top three priorities, and follows execution through — while no sensitive decision happens without your explicit approval.
Built for teams of 1–15 — startups and small businesses
Top three priorities
- 1Close the key client contract
- 2Clear the warehouse bottleneck
- 3Hire a support specialist
Why important work stalls
In most small teams the problem isn't a missing tool — it's that nobody regularly asks "what matters most this week, and who owns it?". Project trackers hold the list but never decide, and the manager burns out between meetings and follow-ups. AI Manager fills exactly that gap: a disciplined weekly cycle that prepares the decision from your team's real data and follows execution to the end.
The weekly management cycle
Six steps that repeat every week, none of them relying on anyone's memory
- 1
Read the situation
Resources, constraints, weekly capacity per person and short team check-ins, recorded and kept current in one place.
- 2
Measure the gap
Your approved 90-day goal is the baseline, and the distance from today to it is measured against defined indicators.
- 3
Propose the week
Three main problems, three priorities, a suggested capacity allocation, risks and what information is missing — each with its reasoning.
- 4
Your approval
Accept, edit or reject the plan. Sensitive decisions come with a structured approval form.
- 5
Execute and track
The approved plan becomes assigned tasks; dependencies, handoffs and real progress are recorded as they happen.
- 6
Compare and learn
At week's end actual results are compared with expectations and written to the decision ledger, so next week's call is better informed.
What you get
The pieces disciplined management actually needs — not a decorative dashboard
A weekly plan with reasons
Three priorities, three problems, a capacity proposal and a risk list; every suggestion is explainable and regenerates when its inputs change.
Authority stays with you
Four configurable authority levels define where the system may act on its own and where it must wait for your written approval.
Auditable tasks
A strict state machine, owners, completion criteria, dependencies, subtasks, handoffs and a full activity timeline.
Decision ledger
Every decision is recorded with its assumptions, expected outcome and actual outcome — a history of what genuinely worked.
Weekly report and alerts
A weekly management report with PDF export, plus in-app notifications and optional SMS for work that has come due.
Isolated workspaces
Each business keeps strictly separate data, and if you run several, they all sit together in one portfolio home.
A clear, configurable authority boundary
The system acts independently only inside the boundary you set; outside it, it waits for you.
Level 1 — Inform
Reports only, changes nothing.
Level 2 — Propose
Drafts plans and tasks; execution needs your approval.
Level 3 — Bounded action
Handles routine, low-risk work inside defined limits.
Level 4 — Sensitive decision
Always requires structured, recorded owner approval.
Honesty instead of false confidence
Every piece of data carries a label: verified fact, owner-reported, assumption, estimate, or missing. Calculations — capacity, budgets, KPIs, due dates — run in deterministic code, not in a language model; the AI interprets and drafts, but never calculates and never changes your business data behind your back.
What it is not
- Not a general chatbot — for open conversation we have Fahmara.
- Not an autonomous CEO; strategic direction is always yours.
- It never spends money, hires, or makes legal calls on its own.
- Not a replacement for full accounting, CRM or inventory systems.
Who it is built for
Two operating modes — you approve which one fits at the start
Startup mode
For early-stage founders: assumptions, experiments and fast learning. The focus is what you need to learn this week.
Organization mode
For managers of running businesses: planning, KPIs, coordination and control against your team's real capacity.
Frequently asked questions
What managers ask before starting
How is this different from project management software?
A project tracker is where you record work; it never decides what matters most. AI Manager starts from your approved goal, your team's real capacity and the current situation, proposes the week's priorities, puts sensitive decisions in front of you for approval, and compares actual results with expectations at week's end.
Does the AI decide instead of me?
No. You set the authority level, and sensitive decisions always require your explicit, recorded approval. The AI analyses the situation and prepares drafts; the choice stays yours.
Where is our business data stored?
Each business gets a strictly isolated workspace with separate data. Deployment on your own dedicated server is also possible; in the demo session we pick the right option based on how sensitive your data is.
How long does setup take?
Initial setup usually takes one to two weeks: completing the business baseline, defining the 90-day goal, and recording the team and their capacity. You can get your first weekly plan in that first week.
What team size does it suit?
The current version is designed for one owner and up to about 15 members per workspace — exactly startup and small-business scale. If you run several businesses, they are managed from one portfolio home.
What does it cost?
Pricing depends on user count, deployment type (cloud or dedicated server) and how much customisation you need. In the first free session we define the scope and give a fixed price with milestone scheduling.
Looking for a Persian chatbot?
Fahmara is our Persian AI assistant with pay-as-you-go pricing.
Start next week with a clear plan
In one short session we walk AI Manager through your own situation and tell you honestly whether it fits your team.
No commitment, no sales pressure