The autopilot assigns by what work actually costs
Most teams assign on gut feel and availability; cost never enters the decision. Infersync captures the real cost of every task from time tracking, hours times each member's rate, and feeds it straight into the autopilot's assignment ranking, so work routes to the best-fit person at the right price and over-budget candidates drop out automatically.
Why teams end up here
- Assignment decisions get made on availability alone; the cost of routing a P1 to a senior engineer instead of a mid-level never enters the equation.
- Cost-per-task gets pieced together in a spreadsheet after delivery, when nobody can act on it.
- Time tracking lives in a separate tool from the work items, so the hours never reconcile with what shipped.
- Nobody knows what a given engineer-hour actually costs the project until the invoice lands.
- When a senior burns budget on routine work a cheaper teammate could own, there's no guardrail that catches it at assignment time.
Cost captured per task from time tracking, then consumed by the AI to assign by cost
Engineers clock in and out on the work items they're already using. Hours times each member's rate is the exact cost per task, no estimates, no separate dashboard. The autopilot assignment ranker reads that cost directly: it scores candidates on skill fit, availability, and inverse cost, and auto-excludes anyone whose rate would push the work over its budget. Cost stops being a report nobody opens and becomes the thing that shapes who gets the work.
Real cost per task from time tracking
Clock in/out and work-item timers capture real hours. Multiplied by each member's rate, that's the exact cost of every task, known, not estimated.
Cost as an input to AI assignment
The assignment ranker scores candidates on skill fit × availability × inverse cost, so the AI routes work to the best-fit person at the right price. Cost shapes the decision, not a dashboard.
Over-budget candidates auto-excluded
Any candidate whose rate would push a task over its budget threshold is dropped from the ranked list before the human sees it. No manual cost discipline required.
Two-way GitHub sync as the source
Issues and PRs from your existing GitHub repos drive the work items the time and cost attach to. No double entry. Edits in Infersync write back to GitHub via webhook.
Per-member rates
Each member has an hourly or daily rate in workspace settings, defaulting to a sensible average if you'd rather not set it per person. That rate is what turns tracked hours into real cost.
Audit trail on assignments
Every assignment the autopilot makes is audit-logged with who-assigned-what-to-whom, so you can see and roll back cost-driven routing decisions.
Steps to get from zero to live
- 1
Connect GitHub
OAuth in 30 seconds. Two-way sync covers issues, PRs, labels, assignees, state, comments.
- 2
Set seat costs
Per-member hourly or daily rate in workspace settings. Defaults to a sensible average if you don't want to set it per person.
- 3
Track time on work items
Engineers clock in/out and run work-item timers. Hours times rate becomes the exact cost per task automatically.
- 4
Let the AI assign by cost
Ask the autopilot to assign work; the ranker uses real cost per task to route it to the best-fit person and skip over-budget candidates.
Available on the Operations plan and up
Time tracking and real cost per task ship on Base (£7/seat/month). The autopilot assignment ranker that consumes cost, including over-budget exclusion, lives on Operations (£15/seat/month). The 14-day free trial unlocks Operations-tier features for all users regardless of plan, so you can see cost-aware assignment against your real GitHub data before committing.
FAQs about the autopilot assigns by what work actually costs
How is the cost of a task actually computed?
It's deterministic: the real hours tracked on a work item (from clock in/out and the work-item timer) multiplied by the assignee's rate (set in workspace settings). That's the exact cost of the task, no estimation, no statistical model. As more time entries land, the figure stays current.
Is there a burn dashboard or cost report I can open?
Cost isn't a separate analytics dashboard. The point is that cost is captured per task from tracked time and consumed by the autopilot to assign by cost, it lives inside the assignment decision rather than in a chart or a PDF. You see the cost on each work item and in the autopilot's preview when it explains a ranking.
What if my team doesn't currently track time?
Use Infersync's native clock in/out and work-item timer with daily reminders, no separate tool needed. Until real hours accumulate, the cost figure is thin, so the AI leans more on skill and availability; precision improves as the team adopts the timer.
What's the over-budget behaviour exactly?
When the autopilot ranks candidates for a new assignment, it computes skill_fit × availability × (1 / cost_per_hour). Any candidate whose rate would push the task over its budget threshold is excluded from the ranked list before the human sees it. The threshold is configured per work item (or inherited from a project default). The reasoning is shown in the autopilot preview step before execution.
Does this work without an AI / LLM key?
The cost capture itself is pure deterministic computation from tracked hours and rates, no LLM needed. The natural-language assignment step uses an LLM; you bring your own Anthropic / OpenAI / Google key (zero token markup from us), or the ranker falls back to deterministic skill + cost + availability ranking without natural-language input.
Can agencies use this to bill clients?
Yes. Because real hours are tracked against work items and each member has a rate, the tracked time gives you defensible hours to bill against. The same data the AI uses to assign by cost is the data you can hand a client as the record of work done.