Frequently asked questions
Schedule-risk management for R&D — CCPM, Monte Carlo, decision gates, and a schedule-aware AI copilot. If your question is not here, email support@delta3coretec.com.
General
What is CritPath AI?
CritPath AI is a web-based schedule-risk platform for R&D programs. It combines real probabilistic scheduling math — Critical Path Method (CPM), Critical Chain (CCPM) with Theory of Constraints, Monte Carlo simulation, and decision gates — with a modern interface and a Claude + Gemini AI copilot that reasons over your actual dependency graph rather than acting as a generic chatbot pinned on top.
Who is CritPath AI for?
It is built for teams running uncertain, milestone-driven research programs: biotech program leaders (VP Development, Program Director) at Series A to C startups; deep-tech founders and CTOs in hardware, fusion, space, battery, semiconductor, and quantum; and principal investigators and program managers on federally funded R&D such as DARPA, ARPA-E, ARPA-H, and NIH grants.
How is it different from the tools I already use?
Legacy schedule-risk tools (Primavera Risk Analysis, Deltek Acumen, Pertmaster) have real Monte Carlo math but cost roughly $10K and up per seat, run on Windows desktop, and are shaped for construction. Modern PM tools (Monday, Asana, Smartsheet, Linear) have great UX but no probabilistic schedule-risk math — their AI flags anomalies, it does not compute a P80 date. CritPath AI is the only product that delivers both the math and the modern UX, plus a schedule-aware AI copilot, at $10 per user per month.
Methodology
What is Critical Chain Project Management (CCPM)?
Critical Chain Project Management is a scheduling method from the Theory of Constraints. Instead of padding every task with hidden safety, CCPM strips task estimates to an aggressive median and pools the protection into shared buffers — a project buffer at the end of the longest resource-feasible chain (the critical chain) and feeding buffers where supporting chains merge in. You then manage the plan by watching buffer consumption rather than individual task due dates.
How is CCPM different from the Critical Path Method (CPM)?
CPM finds the longest chain of dependent tasks ignoring resource contention, and protects each task individually. CCPM (Critical Chain) accounts for resource constraints, removes task-level padding, and protects the whole chain with explicit buffers sized from the project's uncertainty. CritPath AI runs both: a full CPM engine for the network and critical path, and CCPM with Drum-Buffer-Rope for the resource-feasible chain, buffers, and a fever chart.
What is a Monte Carlo schedule simulation and what is a P80 date?
A Monte Carlo schedule simulation runs your network thousands of times, each time drawing a random duration for every task from a three-point (optimistic, most likely, pessimistic) distribution and injecting risk events. The result is a distribution of possible finish dates. A P80 date is the date by which the project finishes in 80 percent of simulations — a far more honest commitment than a single deterministic Gantt date, which is typically only 10 to 20 percent likely. CritPath AI reports P50, P80, and P90 with criticality and tornado sensitivity.
What are project buffers and a fever chart?
A buffer is pooled schedule protection placed at the end of a chain instead of inside each task. A fever chart plots how much of that buffer has been consumed against how much of the chain is complete, mapped to green, yellow, and red zones. It tells you at a glance whether a program is on track or burning protection too fast, which is the core early-warning signal in Critical Chain management.
Does CritPath AI support AACE 132R-23 Level 4?
Yes. CritPath AI implements AACE Recommended Practice 132R-23 Level 4 risk-driven scheduling — three-point estimates, risk registers driving the simulation, contingency derived from the P-levels, and audit trails — so the schedule is defensible for pharma IND programs and federally funded R&D. 21 CFR Part 11 electronic-signature support is on the Enterprise roadmap.
AI copilot & AI employees
What are AI employees?
AI employees are configurable AI agents you set up with their own role and custom skill sets — for example an estimator, a risk analyst, or a status reporter. You assign them to tasks and review their work with a human in the loop. You can build out a whole organization of AI employees alongside your human team to improve productivity, and they are billed by usage rather than per seat.
What can the AI copilot do?
The copilot reasons over your real CPM, TOC, and Monte Carlo results. It can tell you which task is driving your P80 slip, propose three-point estimates, draft risks, generate reports, and answer methodology questions grounded in CPM, CCPM, Monte Carlo, and AACE standards with cited sources — not generic advice. It runs on Claude and Gemini, with no OpenAI lock-in.
What AI is available during the beta?
During the current beta the AI assistance features are on — the methodology coach, AI work-breakdown decompose, AI-generated project reports, and the skill-authoring wizards — so you can learn the product. Autonomous AI-agent runs are turned off during the beta and unlock at general availability.
Pricing & plans
How much does CritPath AI cost?
Pro is $10 per user per month and includes every standard feature: unlimited projects, CPM, Monte Carlo, Theory of Constraints with Drum-Buffer-Rope, decision gates, WSJF, the portfolio dashboard, 2FA/MFA, the audit log, and API access. That is a small fraction of a single legacy schedule-risk license. Enterprise is custom.
How is AI usage billed?
AI employees are billed separately from the seat, by metered usage: your actual LLM token cost plus a platform margin, metered by the hour. You only pay for the AI work you actually run, so a whole org of AI employees scales with use rather than headcount.
Is there a beta or a free trial?
We are running an invite-only beta ahead of general availability, plus a public waitlist you can join from the homepage. Beta organizations get the full scheduling engine and AI assistance features with autonomous agents held until launch.
Security & compliance
Is my data isolated from other organizations?
Yes. CritPath AI is multi-tenant: each signup creates its own organization, and every record is scoped to that organization. Teammates join only by invitation. Your projects, schedules, resources, and AI runs are not visible to any other tenant.
What security features are included?
Accounts use hashed passwords, optional two-factor authentication (TOTP) with one-time recovery codes, and role-based access control across the org-role ladder. An append-only, hash-chained audit log records sensitive actions. Secrets are encrypted at rest.
Do you offer enterprise, on-prem, or HIPAA?
Yes, on the Enterprise plan: an on-prem deployment option, SOC 2 Type II, SAML single sign-on, a HIPAA Business Associate Agreement, a dedicated customer success manager, and custom integrations. 21 CFR Part 11 electronic signatures are on the roadmap.
Getting started
Can I import my existing schedule?
Yes. You can import a spreadsheet (XLSX or CSV) with WBS codes, task names, durations, and predecessors, and CritPath AI builds the project, hierarchy, and dependency network for you. You can also start from a blank project or let the AI decompose a goal statement into a work breakdown.
How do I get access?
Join the waitlist from the homepage. Invited beta organizations get an email link to set up their workspace; the person who signs up becomes the organization owner and can invite their team.
Ready to model your next milestone?
Join the waitlist and we’ll get you into the beta.
Join the waitlist