CEDX Projects · Projects & Delivery
Most project tools tell you what slipped last week. Projects watches the active sprint and says what is about to slip: 32 issues blocked across the demo workspace, a carry-over forecast of 22 pts at close, and the exact issue driving it.
The slip predictor on the Overview does the arithmetic in the open — “If DATA-904 stays blocked, carry forecast rises from 14 → 24 pts (+71%).” You can argue with a forecast that shows its working.
Runs on demo data — the workspace and issue keys in the captures are the seeded demo set.
What it is
Twelve epics on a Q2–Q4 2026 timeline, bar = target window, each with status, priority, progress, points and issue count. Seven are in progress, one is blocked — and the blocked one, the analytics warehouse cutover, renders in red instead of hiding in an average.
The Sprints screen puts the arithmetic on one line: capacity 48 pts, committed 52 pts, completed 19, scope added 6 mid-sprint. Sprint 42 is over-committed by 4 pts and the product says so in orange, not in a post-mortem.
Scope creep gets a number, not a vibe: 6 pts added mid-sprint in week 32, the fourth consecutive sprint with adds, with a guardrail suggestion — cap scope-adds at 4 pts or require a swap. Each alert arrives root-caused: symptom, cause, fix.
Product tour
Not a mockup and not a concept deck. This is what opens at /app/projects.
01 — Overview
251 open issues, 111 in progress, velocity 40 pts over six sprints, 32 blocked and a carry-over forecast of 22 pts — each card sparklines its own trend. Below, the root-caused alerts name the issue and the points at stake: INV-1123 carry risk 86%, Sprint 42 over-committed +6 pts.
02 — Sprints
Twelve sprints tracked, history and future in one table. The active sprint carries 52 pts against 48 pts of capacity with 6 pts of scope added mid-flight — the progress bar reads 37% and the projected carry is 22 pts at the 95th-risk marker.
03 — Triage
Nine items await a decision; the queue suggests routing from component ownership before anything hits the backlog. Each row offers Accept, Route → component, or Defer — with the confidence printed: 91% on the checkout route, 61% on mobile, and duplicates flagged at title similarity ≥ 75%.
04 — Roadmap
The timeline is a target window, not a promise: Q2–Q4 2026, twelve epics, seven in active delivery, one blocked. Under the bars, the same epics as a ledger — status, priority, progress, points, issues, owner, target end — with CSV export of the view you are looking at.
Who runs it
Roles, not references. We have no named customers yet, so nobody in these photographs is quoted, credited or claimed as one.
Runs Sprint 42 off the Sprints screen: 52 committed against 48 of capacity, scope adds argued one at a time, carry forecast on the wall at stand-up.
velocity · 40 pts ±2.2
Watches the blocked column and the people behind it — 32 blocked issues is a staffing fact as much as a process one.
blocked · 32
Owns the twelve epics and says the blocked one out loud. Progress on the timeline is a target window, and the ledger underneath backs it with points and issue counts.
epics · 12 on roadmapThe shape of it
Every figure below is legible in the captures above. Nothing here is a projection of your estate — it is the state of the demo data.
How it runs
Work lands in the intake queue with a suggested route from component ownership and a confidence score. Accept, route, or defer — nothing enters the backlog silently.
The sprint is planned against capacity, not hope: 52 pts committed on 48 pts of capacity is visible the moment it happens, with the over-commit in orange.
The board moves 340 issues through backlog, todo, in progress, review and done, while the slip predictor re-prices carry-over on every change.
Velocity, throughput and the blocked median close the loop. The roadmap's red bar is the conversation the next planning session starts with.
One record
Projects does not keep a parallel copy of the work. The sprint, the issue and the roadmap entry are the same row read at different altitudes.
The capacity, commit-risk scoring and carry forecast behind the Sprints screen live in Sprints — Projects reads them, it does not recompute them.
CEDX IssuesEvery key on the board — INV-1123, DATA-904 — is an issue with labels, SLA clocks and a workflow, not a card-shaped note.
CEDX RoadmapThe twelve epics on the Projects roadmap are initiatives on the portfolio roadmap, with effort and value scores attached there.
CEDX TasksThe follow-ups that are not issues — renewals, requests, checklists — live in Tasks, against the same people and dates.
Limits
Finding this out on the third call is worse for you than reading it here, and worse for us.
Start
Open it right now
Your board, your components
Projects with the rest of it
Questions
Yes. Every screenshot is a capture of the running build and you can open the same build at /app/projects. It runs on demo data, which the captures themselves make legible.
Issue by issue, in the open. The slip predictor scores each committed issue — INV-1123 reads 86% because it is blocked with 7 pts remaining — and sums the expected remainder into the 22 pts carry forecast. The “If DATA-904 stays blocked” card shows the counterfactual arithmetic on the screen.
Visibility first: the Sprints screen prints scope added — 6 pts in week 32 — next to capacity and commitment. The Overview goes further and suggests the guardrail: cap scope-adds at 4 pts or require a swap, after four consecutive sprints with adds.
From component ownership. Each intake row arrives with a suggested component and owner and a confidence score — 91% on the top row — and a human on duty accepts, re-routes or defers. Duplicates are flagged at title similarity ≥ 75% before they reach the backlog.
It is not a separate tracker. The issues on the board are CEDX Issues records with labels, workflows and SLA clocks; Projects is the plan and delivery view over the same rows.
No certification has been issued. What we can evidence about hosting, encryption, tenant isolation and production access is written up on the security page.
Live build, demo data, no card. Then check what your current sprint would score on the commit-risk table.