Mock Room Launcher
选择任意题目立即开始计时模拟(支持语音输入、7大岗位结构化草稿纸、可视化系统架构板、Bar-Raiser追问与逐句改写评分)。
Legacy Logistics: 40-Warehouse LLM + ERP 30-60-90 Decomposition
A Fortune 500 logistics client has 40 regional warehouses running three incompatible on-prem ERPs (SAP, Oracle, and custom AS/400 CSV dumps). Dispatchers spend 4 hours/day manually reconciling delayed shipments from PDF manifests and emails. You are the lead FDE deployed on-site on Monday. Decompose the problem, design the production architecture, and lay out your 30-60-90 delivery plan.
Production AI Accuracy Drop (94% → 78%) Pre-Board Demo Triage
You deployed an agentic contract-review assistant at a flagship enterprise customer. Three weeks after launch—and 48 hours before the customer CEO demos it to their Board—their QA lead escalates that accuracy dropped from 94% in staging to 78% in production. Walk me through your live triage, root-cause isolation, and customer communication.
$3M ARR Bespoke Client Integration vs. Platform Abstraction
Your top $3M ARR defense-manufacturing client demands a custom real-time shop-floor scheduling module with proprietary SAP BAPI hooks that no other customer uses. Building it bespoke will take 2 FDEs for 6 weeks and risk creating a permanent code fork. Core Engineering refuses to touch it for 2 quarters. What do you decide and how do you execute?
Offline-Resilient Industrial Multimodal Field Inspection App
Field technicians inspect remote wind turbines and substations with spotty 3G/offline connectivity. They need an AI copilot on iPad/rugged Android that inspects photos/audio notes, flags safety hazards against 5,000-page OEM manuals, and syncs work orders when back online. Architect the edge-to-cloud system.
3 Healthcare Design Partners → 80% Reusable Product Core
You are FDPM for a new Clinical Revenue-Cycle AI product. Three early design partners want completely different things: Hospital A wants EHR claim denial appeals; Hospital B wants prior-auth fax triage; Hospital C wants nurse coding copilot. You only have 1 core squad and 3 FDEs. How do you structure discovery and turn this into an 80% reusable product?
Pilot-to-$1.5M Annual Contract CFO ROI Defense
Your 90-day pilot automating supply-chain vendor claims went live with glowing operator feedback (4.6/5 CSAT). Now the client CFO is stalling on the $1.5M/year enterprise contract because "AI demos always look good, show me hard P&L savings and risk control." Build the conversion case and rollout structure.
Navigating FDE vs. Client Security vs. Core Eng Priorities
Tell me about a time—or walk through how you handle a scenario—where Field Engineers wanted to ship an urgent custom workaround for a flagship customer, Client InfoSec blocked data access for 3 weeks, and Core Engineering pushed back on every dependency. How do you align all three without burning trust?
Pricing & Packaging an Outcome-Based Forward-Deployed AI Agent
You built an enterprise AI Agent for B2B freight brokerage & exception resolution. Historically your company charged seat-based SaaS + fixed implementation fee. Now automation cuts customer seats by 40%, threatening NRR. Design the new pricing, packaging, and forward-deployed services model.
Enterprise RAG + Agent Eval Harness Design
Your enterprise knowledge assistant answers employee policy, IT, and sales questions across Slack, Confluence, Jira, and Google Drive. Users complain answers are sometimes outdated, ignore permissions, or sound confident when wrong. Design the end-to-end Eval Harness (offline + online) before your next major release.
Cutting Agent Inference COGS 60% Without Quality Regression
Your multi-step AI coding/research agent has strong retention, but inference spend just hit 52% of revenue (dragging gross margin down to 40%). The CFO gives you 6 weeks to cut inference COGS by 60% while holding task completion rate flat within 1%. Design your optimization roadmap.
HITL & Failure Recovery UX for Non-Deterministic Agents
You are launching an AI Financial Ops Agent that can read vendor invoices, query ERP balances, draft journal entries, and initiate payments up to $50K. Users love automation on routine invoices, but one hallucinated wire transfer will get your product ripped out. Design the autonomy levels, Human-in-the-Loop UX, and failure recovery.
Prompt + RAG vs. Fine-Tune vs. Custom Model Build-or-Buy
A B2B legal-tech startup wants to improve contract redlining quality against each customer’s proprietary playbook. Your VP of Eng pitches fine-tuning an open-weights 70B model per customer; your Lead AI Engineer wants advanced RAG + frontier APIs. As AI PM, frame the build-vs-buy and architecture decision.
0→1 Product Sense: Turning Support Calls into Prioritized Eng Tickets
Design a product for B2B SaaS teams that turns messy customer support calls, Slack threads, and tickets into prioritized engineering bugs and feature insights. Walk through users, pain points, product design, differentiation, and metrics.
Root-Cause Diagnosis: Weekly Active Teams -14% WoW Post-Onboarding Redesign
You are PM for a collaborative B2B workflow platform. On Monday morning, Weekly Active Teams dropped 14% week-over-week—two weeks after you shipped a redesigned self-serve onboarding flow. Walk me through how you diagnose the drop and decide your next move.
Metrics Hierarchy & Guardrails for an AI Consultant Marketplace
You run a two-sided marketplace matching Fortune 1000 companies with vetted AI Transformation Consultants & FDEs for 4-to-12 week sprints. Define the North Star metric, ecosystem health metrics, guardrail metrics, and how you detect marketplace gaming.
Enterprise Vertical Workflow Suite vs. Horizontal Developer API
Your company built a breakthrough document & workflow intelligence engine. Currently 70% of revenue comes from a horizontal developer API ($0.02/page), while sales keeps asking for a full vertical SaaS suite for Insurance Underwriting priced at $250K/year. As Product Strategy lead, should we double down on the API, build the vertical suite, or sequence both?
Enterprise Onboarding Time (18d → 47d) Under 3x Bookings Growth
Enterprise bookings tripled this quarter—great news! However, median customer onboarding time blew up from 18 days to 47 days, creating a $4.2M backlog of unactivated ARR and rising pre-launch churn. You are Strategy & Ops Lead. Diagnose the bottleneck and cut onboarding back under 20 days within 8 weeks.
European Managed Services Expansion: Capacity & Unit Economics
Your US AI Data & Managed Operations business runs at 58% gross margin using a blended onshore/nearshore workforce. Leadership wants to launch in Germany and the UK next quarter to support 4 large European enterprise contracts under strict EU data residency. Model the capacity, unit economics, and operational launch plan.
Fixing Broken Incentives: Launch Velocity vs. 2x 6-Month Churn
Six months ago, you shifted Deployment & Ops leads’ bonuses to reward "Days-to-Launch" speed. Launch speed improved 35%—but 6-month customer churn doubled and support escalations spiked 3x. Diagnose the incentive failure and redesign the operating model and compensation scorecards.
COO Monday WBR Dashboard (7 Metrics & Red-Flag Triggers)
You are Chief of Staff / Ops Lead for a fast-growing $40M ARR AI Deployment & Operations org (120 FDEs, Ops managers, and Support engineers). The COO asks you to design the Monday Weekly Business Review (WBR) scorecard: pick the exact 7 metrics, explain why each earns its slot, and define the red-flag escalation triggers.
GA-3 Weeks War Room: Legal Data Residency + 2x Peak Load Risk
You are Lead Technical Program Manager for an enterprise AI suite scheduled for General Availability in 3 weeks—anchored to an annual user conference keynote. Today: (1) Legal flags an EU data-residency subprocessor gap blocking 30% of launch pipeline, and (2) Load testing fails at 60% of projected keynote traffic. Run the 3-week war room.
Multi-Region Platform Migration Across 6 Eng Teams + 10 Design Partners
Your company must migrate 10 live enterprise design partners and 6 internal engineering teams from a legacy single-tenant V1 pipeline onto a new multi-tenant V2 agent runtime within 16 weeks—with zero customer downtime and strict backward API compatibility. Design the program architecture, dependency map, and migration sequence.
Pushing Back on Mid-Quarter VP Demo Scope Creep
Halfway through a tightly scoped Q3 reliability & core latency program, a senior VP sees a competitor demo and asks your team to squeeze in a flashy new voice-agent feature for an upcoming investor meeting in 3 weeks. Describe how you evaluate the request, push back or negotiate options, and protect team morale.
Scaling Delivery Governance from 40 to 250 Builders Without Bureaucracy
Eighteen months ago, 40 engineers and FDEs shipped fast via informal Slack chats. Now the org is 250 people across 3 time zones, and launches are colliding, dependencies break silently, and engineers complain about "too many status meetings." Design the minimal operating rhythm & governance system that scales.
B2B SaaS + Services: Revenue +25% YoY While EBITDA Margin Fell 22% → 9%
Your client is a $120M B2B Industrial Analytics software + implementation services company. Revenue grew 25% YoY, yet EBITDA margin collapsed from 22% to 9% over 4 quarters. The CEO asks you to find out why and present a turnaround plan to restore 20%+ EBITDA within 12 months.
Market Sizing: Annual AI Workflow Software Spend in US Mid-Market Logistics
Estimate the total addressable annual software spend (TAM & SAM) for AI-powered dispatch & back-office workflow automation across US mid-market logistics and freight companies (50–1,000 employees). Show your structure, assumptions, math, and sanity check.
Healthcare Diagnostics Leader: AI Triage Build vs. Buy vs. Partner
A $4B public healthcare diagnostics company processes 250M lab orders/year. Management wants to deploy generative AI + multimodal triage across prior-auth and abnormal result follow-ups within 12 months. Should they Build in-house, Buy a specialized startup, or Partner with a frontier AI/FDE vendor? Structure the decision.
60-Second Pyramid Principle Synthesis to a Skeptical CEO
Imagine you just spent 3 weeks investigating why a global B2B tech company’s new AI Operations platform has 85% pilot sign-ups but only 22% daily active usage after Month 2. You have 60 seconds with an impatient CEO walking to the elevator. Deliver the Pyramid Principle synthesis—and defend against rapid-fire pushback.