Trending Prompts

Implement comprehensive model evaluation and validation frameworks with proper metrics and statistical analysis. Classification metrics: 1. Accuracy: correct predictions / total predictions, baseline comparison, stratified sampling. 2. Precision: true positives / (true positives + false positives), minimize false alarms. 3. Recall (Sensitivity): true positives / (true positives + false negatives), capture all positive cases. 4. F1-score: harmonic mean of precision and recall, balanced metric for imbalanced datasets. Regression metrics: 1. Mean Absolute Error (MAE): average absolute differences, interpretable units, robust to outliers. 2. Root Mean Square Error (RMSE): penalizes large errors, same units as target variable. 3. R² (coefficient of determination): explained variance, 1.0 = perfect fit, negative = worse than mean. Advanced evaluation: 1. ROC-AUC: area under ROC curve, threshold-independent, >0.9 excellent performance. 2. Precision-Recall curve: imbalanced datasets, focus on positive class performance. 3. Confusion matrix: detailed error analysis, class-specific performance, misclassification patterns. Cross-validation strategies: 1. Stratified K-fold: maintain class distribution, k=5 or k=10, repeated CV for stability. 2. Time series validation: walk-forward, expanding window, respect temporal dependencies. 3. Leave-one-out: small datasets, computationally expensive, unbiased estimates. Statistical significance: 1. Paired t-test: compare model performance, statistical significance p<0.05. 2. Bootstrap sampling: confidence intervals, performance stability assessment. 3. McNemar's test: classifier comparison, statistical hypothesis testing. Business metrics integration: ROI calculation, cost-benefit analysis, domain-specific targets, A/B testing framework for production validation.

Implement AI safety measures including robustness testing, adversarial attack detection, and defense mechanisms for secure AI systems. Adversarial attacks: 1. FGSM (Fast Gradient Sign Method): single-step attack, epsilon perturbation, white-box scenario. 2. PGD (Projected Gradient Descent): iterative attack, stronger than FGSM, constrained optimization. 3. C&W attack: optimization-based, minimal distortion, confidence-based objective function. Defense mechanisms: 1. Adversarial training: include adversarial examples in training, robustness improvement, min-max optimization. 2. Defensive distillation: temperature scaling, smooth gradients, gradient masking prevention. 3. Input preprocessing: denoising, compression, randomized smoothing, transformation-based defenses. Robustness evaluation: 1. Certified defenses: mathematical guarantees, interval bound propagation, certified accuracy. 2. Empirical robustness: attack success rate, perturbation budget analysis, multiple attack types. 3. Natural robustness: corruption robustness, out-of-distribution generalization, real-world noise. Detection methods: 1. Statistical tests: input distribution analysis, feature statistics, anomaly detection. 2. Uncertainty quantification: prediction confidence, ensemble disagreement, Bayesian approaches. 3. Intrinsic dimensionality: manifold learning, adversarial subspace detection. Safety frameworks: 1. Alignment research: reward modeling, human feedback, value alignment, goal specification. 2. Interpretability: decision transparency, explanation generation, bias detection. 3. Monitoring systems: drift detection, performance degradation, safety constraints. Red teaming: systematic testing, failure mode discovery, stress testing, security assessment protocols, continuous monitoring for emerging threats and vulnerabilities.

Implement MLOps practices for scalable machine learning deployment, monitoring, and lifecycle management. MLOps pipeline stages: 1. Data versioning: DVC (Data Version Control), data lineage tracking, feature store management. 2. Model training: automated retraining, hyperparameter optimization, experiment tracking with MLflow. 3. Model validation: A/B testing, shadow deployments, performance regression testing. 4. Deployment: containerized models (Docker), API serving (FastAPI, Flask), batch prediction jobs. Model serving strategies: 1. REST API: synchronous predictions, load balancing, auto-scaling based on request volume. 2. Batch inference: scheduled jobs, distributed processing with Spark, large dataset processing. 3. Real-time streaming: Kafka integration, low-latency predictions (<100ms), edge deployment. Monitoring and observability: 1. Data drift detection: statistical tests, distribution comparison, feature drift alerts. 2. Model performance: accuracy degradation monitoring, prediction confidence tracking. 3. Infrastructure metrics: CPU/memory usage, request latency, error rates, throughput monitoring. ML infrastructure: 1. Feature stores: centralized feature management, real-time/batch serving, feature lineage. 2. Model registry: versioning, metadata storage, deployment approval workflows. 3. Experiment tracking: hyperparameter logging, metric comparison, reproducible results. CI/CD for ML: 1. Automated testing: unit tests for preprocessing, integration tests for pipelines. 2. Model validation: holdout testing, cross-validation, business metric validation. Tools: Kubeflow for Kubernetes, SageMaker for AWS, Azure ML, Google AI Platform, target deployment time <30 minutes.

Develop a retention script for cancellation requests. Strategy: 1. Acknowledge their decision respectfully. 2. Ask open-ended question about reason. 3. Listen actively to their concerns. 4. Offer targeted solutions (pause subscription, discount, feature upgrade). 5. Highlight value they'll lose. 6. Present limited-time retention offer. 7. Make cancellation easy if they insist. 8. Request feedback for improvement. Never be pushy or manipulative.

Develop local marketing strategies with hyperlocal targeting and community engagement for location-based businesses. Local SEO optimization: 1. Google My Business: complete profile, regular posts, photo updates, review management, Q&A monitoring. 2. Local citations: NAP consistency (Name, Address, Phone), directory submissions, industry-specific listings. 3. Location pages: unique content per location, local keywords, maps integration, contact information. Hyperlocal targeting: 1. Geographic targeting: radius targeting, zip code level, neighborhood focus, competitor location analysis. 2. Local keywords: 'near me' searches, city + service, neighborhood names, local landmarks. 3. Community involvement: local events, sponsorships, partnerships, charitable activities, local news. Community engagement: 1. Local partnerships: cross-promotion, referral programs, joint events, business associations. 2. Event marketing: community events, grand openings, seasonal celebrations, workshop hosting. 3. Local influencers: micro-influencers, community leaders, local celebrities, customer advocates. Digital local marketing: 1. Social media: location tagging, local hashtags, community groups, neighborhood targeting. 2. Local advertising: Facebook local awareness ads, Google Local Services ads, Nextdoor advertising. 3. Review management: Google reviews, Yelp, Facebook reviews, response strategy, reputation building. Traditional local marketing: 1. Print advertising: local newspapers, magazines, direct mail, flyers, community bulletins. 2. Radio/local TV: sponsorships, talk show appearances, community calendar listings. 3. Outdoor advertising: billboards, transit advertising, local signage, vehicle wraps. Measurement: foot traffic analysis, local search rankings, review sentiment, community engagement metrics, local market share assessment for neighborhood dominance.

Deliver personalized demos that convert. Pre-demo research (30 mins): 1. Company website, recent news, LinkedIn. 2. Prospect's role, responsibilities, pain points from discovery. 3. Current tools they use (from conversation or SimilarWeb). Demo customization: 1. Use prospect's company name in demo environment. 2. Import sample data relevant to their industry. 3. Show workflow that mirrors their process. 4. Address specific pain points discovered. 5. Skip features they don't care about. Opening: 'Based on our conversation, I've customized this to show...' Throughout: ask confirming questions ('Is this how you currently do it?'). End: clear CTA and next steps. Follow-up: personalized recap email with screenshots.

Execute international marketing strategies with cultural adaptation and localization for global market expansion. Market research and entry: 1. Market analysis: market size, competition, regulatory environment, cultural factors, economic conditions. 2. Entry strategy: direct investment, partnerships, distributors, licensing, franchising, e-commerce platforms. 3. Competitive landscape: local competitors, international brands, pricing strategies, market positioning. Cultural adaptation: 1. Localization strategy: language translation, cultural nuances, local customs, religious considerations. 2. Visual adaptation: color psychology, imagery preferences, design aesthetics, cultural symbols. 3. Product adaptation: feature modifications, packaging changes, sizing adjustments, regulatory compliance. Digital marketing localization: 1. Website localization: language translation, currency conversion, local payment methods, cultural design. 2. Search marketing: local keyword research, search engines preferences (Baidu, Yandex), local SEO. 3. Social media: platform preferences, content adaptation, local influencers, cultural communication styles. Global campaign management: 1. Brand consistency: global brand guidelines, local adaptation parameters, approval processes. 2. Campaign coordination: timing considerations, cultural events, seasonal differences, local holidays. 3. Budget allocation: market prioritization, investment levels, performance expectations, ROI targets. Legal and compliance: 1. Advertising regulations: content restrictions, disclosure requirements, competitive claims, data privacy. 2. Data protection: GDPR, local privacy laws, data localization, consent management. 3. Intellectual property: trademark protection, copyright compliance, brand usage rights. Performance measurement: market-specific KPIs, cross-market comparison, cultural performance factors, localization ROI analysis for continuous optimization and expansion planning.

Develop optimal pricing strategy through research and testing. Pricing models: 1. Freemium: free tier + paid upgrades (good for viral/network effects). 2. Tiered: good/better/best packages (most common for SaaS). 3. Usage-based: pay per use/seat/transaction (aligns cost with value). 4. Flat rate: single price (simple but leaves money on table). Research methods: 1. Van Westendorp Price Sensitivity Meter (survey method). 2. Conjoint analysis: test feature/price combinations. 3. Competitor benchmarking: position relative to alternatives. 4. Customer interviews: value perception and willingness to pay. Testing approaches: 1. A/B testing: different prices to new customers. 2. Landing page tests: measure conversion at various price points. 3. Cohort analysis: retention by price paid. Optimization: raise prices annually for new customers, grandfather existing ones. Monitor churn rate changes after price increases.

Optimize e-commerce marketing funnels with conversion strategies and customer acquisition tactics for online retail. E-commerce funnel optimization: 1. Traffic generation: SEO, PPC, social media, email marketing, affiliate partnerships, influencer collaborations. 2. Product discovery: site search optimization, category navigation, filtering, personalized recommendations. 3. Conversion optimization: product pages, cart abandonment, checkout process, payment options, trust signals. Product marketing: 1. Product descriptions: benefit-focused copy, SEO optimization, social proof integration, technical specifications. 2. Visual merchandising: high-quality images, 360-degree views, zoom functionality, video demonstrations. 3. Pricing strategy: competitive analysis, dynamic pricing, promotional offers, bundle pricing, psychological pricing. Cart abandonment recovery: 1. Email sequences: immediate reminder (1 hour), incentive offer (24 hours), last chance (72 hours). 2. Retargeting ads: dynamic product ads, cross-platform remarketing, personalized messaging. 3. Exit-intent popups: discount offers, free shipping, chat support, newsletter signups. Customer acquisition: 1. Paid advertising: Google Shopping ads, Facebook catalog ads, Instagram shopping, Amazon advertising. 2. Content marketing: buying guides, product comparisons, how-to content, user-generated content. 3. Social commerce: Instagram Shopping, Facebook Shop, Pinterest Product Rich Pins, TikTok Shopping. Customer lifecycle: 1. First-time buyers: welcome offers, product education, support resources, review requests. 2. Repeat customers: loyalty programs, exclusive offers, early access, personalized recommendations. 3. VIP customers: premium support, exclusive products, special events, referral incentives. Analytics and optimization: conversion rate tracking, customer lifetime value, average order value, return on ad spend (ROAS), cohort analysis for sustainable growth.

Generate natural speech with ElevenLabs. API usage: 1. Choose voice from library. 2. Adjust stability and clarity. 3. Stream audio for low latency. 4. Voice cloning from samples. 5. Multiple languages support. 6. Emotion and style control. 7. SSML for pronunciation. 8. Webhook for long-form content. Implement audio caching and use websocket for real-time streaming.

Optimize AI models for edge deployment with mobile inference, model compression, and real-time processing constraints. Model compression techniques: 1. Quantization: FP32 to INT8, post-training quantization, quantization-aware training. 2. Pruning: weight pruning, structured pruning, magnitude-based pruning, gradual sparsification. 3. Knowledge distillation: teacher-student training, soft targets, temperature scaling. Mobile optimization: 1. Model size constraints: <10MB for mobile apps, <100MB for edge devices. 2. Inference optimization: ONNX runtime, TensorFlow Lite, Core ML for iOS deployment. 3. Hardware acceleration: GPU inference, Neural Processing Units (NPU), specialized chips. Deployment frameworks: 1. TensorFlow Lite: mobile/embedded deployment, delegate acceleration, model optimization toolkit. 2. PyTorch Mobile: C++ runtime, operator support, optimization passes. 3. ONNX Runtime: cross-platform inference, hardware-specific optimizations. Real-time constraints: 1. Latency requirements: <100ms for interactive applications, <16ms for real-time video. 2. Memory constraints: RAM usage minimization, model partitioning, streaming inference. 3. Power efficiency: battery optimization, model scheduling, dynamic frequency scaling. Edge computing scenarios: 1. Computer vision: real-time object detection, image classification, pose estimation. 2. Natural language: on-device speech recognition, text classification, language translation. 3. IoT applications: sensor data processing, anomaly detection, predictive maintenance. Performance monitoring: 1. Inference speed: frames per second, latency percentiles, throughput measurement. 2. Accuracy preservation: model accuracy after compression, A/B testing, quality metrics. 3. Resource utilization: CPU/GPU usage, memory consumption, power draw monitoring, thermal management for sustained performance.

Implement graph neural networks for social network analysis, knowledge graphs, and relational data modeling. Graph fundamentals: 1. Graph representation: adjacency matrix, edge list, node features, edge attributes. 2. Graph types: directed/undirected, weighted/unweighted, temporal, heterogeneous graphs. 3. Graph properties: degree distribution, clustering coefficient, path length, centrality measures. GNN architectures: 1. Graph Convolutional Networks (GCN): spectral approach, Laplacian matrix, localized filters. 2. GraphSAGE: inductive learning, neighbor sampling, mini-batch training on large graphs. 3. Graph Attention Networks (GAT): attention mechanism, node importance weighting, multi-head attention. Message passing: 1. Aggregation functions: mean, max, sum, attention-weighted aggregation. 2. Update functions: neural networks, gated updates, residual connections. 3. Multi-layer propagation: information propagation, over-smoothing prevention, layer normalization. Applications: 1. Node classification: user categorization, protein function prediction, document classification. 2. Graph classification: molecular properties, social network analysis, fraud detection. 3. Link prediction: friendship recommendation, drug-target interaction, knowledge graph completion. Social network analysis: 1. Community detection: modularity optimization, label propagation, community structure analysis. 2. Influence analysis: information diffusion, viral marketing, opinion dynamics modeling. 3. Centrality measures: betweenness, closeness, eigenvector centrality, PageRank algorithm. Implementation: PyTorch Geometric, DGL (Deep Graph Library), graph data loaders, mini-batch sampling, GPU acceleration for large graphs, scalability considerations for million-node networks.

Optimize manuscript for peer review success. IMRAD structure: 1. Introduction: establish importance, review relevant literature, state hypotheses clearly. 2. Methods: detailed enough for replication, justify choices, report deviations from protocol. 3. Results: report findings objectively, use appropriate statistics, include effect sizes and confidence intervals. 4. Discussion: interpret findings, acknowledge limitations, suggest future research. Additional sections: abstract (250 words), keywords, references, figures/tables. Pre-submission: 1. Check journal fit: scope, impact factor, open access policies. 2. Follow journal guidelines exactly: formatting, word limits, reference style. 3. Get colleague reviews, especially from methodologists. Cover letter: highlight novelty and importance, suggest reviewers, declare conflicts of interest. Response to reviewers: address each comment systematically, thank reviewers, clarify but don't argue defensively. Track citations and altmetrics post-publication.

Run focused 5-day design sprints to solve big product challenges. Day 1 (Map): 1. Define long-term goal and sprint questions. 2. Map customer journey from start to finish. 3. Ask 'How Might We' questions and collect notes. 4. Target specific part of journey for sprint focus. Day 2 (Sketch): 1. Lightning demos of inspiring solutions. 2. Four-step sketching: notes, ideas, crazy 8s, solution sketch. Day 3 (Decide): 1. Present solution sketches anonymously. 2. Heat map voting and feedback. 3. Storyboard winning solution for prototype. Day 4 (Prototype): 1. Build realistic prototype (InVision, Figma). 2. Write realistic content, not lorem ipsum. Day 5 (Test): 1. 5 user interviews with prototype. 2. Document learnings and next steps. Team: 7 or fewer people, including decision maker. Outcome: validated or invalidated hypothesis with user evidence.

Diversify an ELA curriculum with culturally responsive texts. Audit Current Curriculum: Analyze current book list for author diversity, character representation, and cultural perspectives. Goal: ensure texts serve as 'windows' (seeing into others' experiences) and 'mirrors' (reflecting students' own identities). Selection Criteria: 1. Authenticity: written by authors from the culture being represented. 2. Complexity: avoids stereotypes and presents nuanced characters. 3. Relevance: connects to students' lives and contemporary issues. Example additions: replace a classic with a contemporary work by a BIPOC author (e.g., 'The Hate U Give' by Angie Thomas, 'There There' by Tommy Orange). Pairings: pair a canonical text with a counter-narrative (e.g., 'The Great Gatsby' with 'Passing'). Involve students in the selection process.

Develop franchise model. Components: 1. Proven concept and unit economics. 2. Franchise agreement (legal document). 3. Franchise fee structure (initial + ongoing royalties). 4. Training program for franchisees. 5. Operations manual (detailed SOPs). 6. Marketing support and brand guidelines. 7. Territory rights. 8. Quality control and audits. Requires FDD (Franchise Disclosure Document). Scale through others' capital. Maintain brand consistency.

Design a VR lesson for a world history class using Google Expeditions or similar platform. Objective: Students will identify key architectural features of the Colosseum and Roman Forum. Pre-VR Activity (10 mins): Introduce key vocabulary (arch, aqueduct, forum) and provide historical context. VR Experience (20 mins): 1. Guide students through a 360-degree tour of the Colosseum. 2. Pause at points of interest, asking questions ('What events took place here?' 'How does the architecture support a large crowd?'). 3. Move to the Roman Forum, have students identify different building types. Post-VR Activity (15 mins): Students write a 'postcard' from ancient Rome describing what they saw, or work in groups to build a model of a Roman structure.

Run effective sprint planning and backlog refinement sessions. Backlog grooming (weekly, 1 hour): 1. Review upcoming stories for clarity and completeness. 2. Add acceptance criteria and designs. 3. Estimate story points (Fibonacci sequence: 1, 2, 3, 5, 8). 4. Identify dependencies and blockers. 5. Split large stories (>8 points) into smaller ones. Sprint planning (every 2 weeks, 2 hours): 1. Review sprint goal and team velocity. 2. Select stories totaling team's capacity. 3. Discuss implementation approach for complex stories. 4. Confirm Definition of Ready for all selected stories. 5. Create tasks and assign owners. Velocity tracking: average story points completed over last 3 sprints. Buffer: reserve 20% capacity for bugs and urgent items. Tools: Jira, Azure DevOps, Linear for story management.

Design an engaging 90-minute PD session on a new instructional strategy. Agenda: 1. Why (10 mins): Start with research or data showing the need for the strategy. 2. What (20 mins): Clearly explain and model the strategy. Show a video of it in action. 3. How (30 mins): Active engagement. Have teachers try the strategy themselves (e.g., plan a short lesson segment using it). 4. What If (15 mins): Facilitate a discussion about potential challenges and solutions for implementation in their own classrooms. 5. Now What (15 mins): Teachers set a specific goal for how they will try the strategy in the next week. Provide a resource handout. Avoid 'sit and get'; prioritize active learning and collaboration.

Compare visual outputs from multiple SD models. Layout: 1. One prompt input that sends to SDXL, SD1.5, and Playground v2. 2. 3-column grid showing generated images. 3. Metadata overlay showing seed, sampler, and CFG scale. 4. 'Download All' button. 5. History sidebar of previous generations.

Create a luxurious Art Deco geometric pattern. Elements: 1. Symmetrical geometric shapes (chevrons, sunbursts, zigzags). 2. Metallic gold and black color palette with emerald accents. 3. High contrast and sharp lines. 4. Repeating motifs inspired by 1920s architecture. 5. Seamless tiling for wallpaper or textile use. Style: opulent, sophisticated, glamorous. Use vector graphics for scalability. Perfect for luxury branding, interior design, or fashion applications.

Implement navigation in React Native apps. Patterns: 1. Stack navigator for hierarchical screens. 2. Tab navigator for main sections. 3. Drawer navigator for side menu. 4. Deep linking and universal links. 5. Screen transitions and gestures. 6. Nested navigators composition. 7. Authentication flow routing. 8. Persistent navigation state. Use React Navigation v6 with TypeScript for type-safe routes and implement header customization.

Create a resumable PWA using Qwik's unique architecture. Requirements: 1. Instant interactivity with 0 hydration. 2. Service worker for offline capability. 3. Lazy-load components on interaction. 4. Streaming SSR with early hints. 5. Smart prefetching based on viewport. 6. App shell pattern for navigation. 7. Push notifications integration. 8. Install prompt with custom UI. Use Qwik City for routing and resumability for instant TTI even on slow networks.

Build modern iOS apps with SwiftUI. Components: 1. View protocol for custom views. 2. @State and @Binding for local state. 3. @ObservedObject for external state. 4. List with ForEach for collections. 5. Navigation with NavigationStack. 6. Async/await for data loading. 7. Custom view modifiers. 8. Animations with withAnimation. Use Combine for reactive programming and implement dark mode support with @Environment.

Ensure intervention delivery matches intended protocol. Fidelity dimensions (NIH BCC): 1. Design fidelity: intervention based on theory and prior evidence. 2. Training fidelity: standardized training for intervention providers. 3. Delivery fidelity: intervention delivered as intended. 4. Receipt fidelity: participants receive and understand intervention. 5. Enactment fidelity: participants use skills in real life. Monitoring strategies: 1. Session checklists: key components delivered (yes/no checklist). 2. Audio/video recording: sample sessions reviewed by independent raters. 3. Participant feedback: exit interviews about intervention components received. 4. Provider self-report: reflection on session delivery and challenges. Assessment tools: 1. Fidelity rating scales: Likert scales for component quality/adherence. 2. Time and motion studies: duration of intervention components. 3. Competence measures: how skillfully intervention delivered. Reporting: describe fidelity monitoring plan in protocol, report actual fidelity in results, discuss implications of low fidelity for interpretation.

Craft compelling stories using proven narrative structures and character development. Three-act structure: Act 1 (25%): setup, introduce character, inciting incident. Act 2 (50%): rising action, obstacles, midpoint crisis. Act 3 (25%): climax, resolution, denouement. Hero's journey: ordinary world → call to adventure → mentor → threshold → tests → ordeal → reward → return transformed. Character development: 1. Want vs. need: surface desire vs. deeper requirement for growth. 2. Internal conflict: contradictory motivations, fears, beliefs. 3. Character arc: how they change from beginning to end. 4. Dialogue voice: unique speech patterns, vocabulary, rhythm. Story elements: 1. Hook: compelling opening that raises questions. 2. Stakes: what character stands to gain/lose. 3. Pacing: balance of action, dialogue, description. 4. Theme: underlying message about human condition. Genre conventions: romance (meet-cute, obstacles, happy ending), mystery (red herrings, clues, revelation), thriller (ticking clock, danger escalation). Writing process: outline → first draft → character consistency check → dialogue polish → final edit.

Optimize retail merchandising. Strategies: 1. Product placement (eye level, end caps). 2. Planogram optimization. 3. Seasonal displays and themes. 4. Cross-merchandising complementary products. 5. Signage and pricing clarity. 6. Inventory visibility. 7. Store traffic flow design. 8. Impulse buy positioning. Use data on sales per square foot. Test and iterate layouts. Visual appeal matters. Balance bestsellers with discovery.

Sell on value, not features. Discovery questions for value: 1. 'What's the cost of the current problem?' (time, money, opportunity). 2. 'What happens if you don't solve this?' (quantify downside). 3. 'How would solving this impact the business?' (revenue increase, cost reduction, risk mitigation). Calculate value together: Current cost: 'You mentioned 3 people spend 10 hours/week on manual reporting, that's 1,560 hours/year. At $50/hour, that's $78k annually.' Solution value: 'Our automation reduces this to 2 hours/week, saving $65k/year.' ROI pitch: '$65k saved, our solution is $30k/year, that's 2.2x ROI and 5.5-month payback.' Compare to alternatives: status quo cost vs. solution cost. Document in mutual plan or proposal. Align pricing to value (if $65k saved, $30k fee is justified). Ask: 'Does that ROI make sense for your business?' Makes price objections irrelevant.

Create a multi-page application with Parcel's zero-config approach. Setup: 1. Multiple HTML entry points. 2. Automatic code splitting per page. 3. Shared chunks for common dependencies. 4. Hot module replacement. 5. Image optimization and resizing. 6. PostCSS and Sass support out-of-box. 7. Environment variable injection. 8. Production builds with tree-shaking. No webpack config needed. Use @parcel/transformer-typescript and implement service worker for offline support.

Maintain research integrity and prevent scientific misconduct. Types of misconduct: 1. Fabrication: making up data or results. 2. Falsification: manipulating research processes or changing results. 3. Plagiarism: using others' ideas without proper attribution. 4. Questionable research practices: p-hacking, selective reporting, inappropriate authorship. Prevention strategies: 1. Research integrity training: responsible conduct of research courses. 2. Data management: audit trails, version control, shared databases. 3. Supervision: regular meetings, progress reviews, co-analysis of data. 4. Institutional culture: open discussion of ethical dilemmas, reporting mechanisms. Detection methods: 1. Statistical screening: digit analysis, impossible values, too-perfect distributions. 2. Image analysis: duplicated images, inappropriate manipulation. 3. Text analysis: plagiarism detection software. 4. Peer review: careful examination of methods and results. Response protocols: 1. Investigate allegations promptly and fairly. 2. Protect whistleblowers from retaliation. 3. Collaborate with journals for corrections or retractions. Restoration: focus on education and prevention rather than punishment alone.

Access multiple LLMs via OpenRouter. Benefits: 1. Single API for 50+ models. 2. Cost comparison across models. 3. Fallback to alternative models. 4. Real-time model availability. 5. Usage analytics dashboard. 6. OpenAI-compatible API. 7. Free models available. 8. Model routing based on performance. Switch models without code changes. Monitor costs and reliability.

Develop culturally sensitive creative campaigns for global markets while maintaining brand consistency. Cultural research framework: 1. Color significance: red = luck (China) vs. danger (West), white = purity vs. mourning. 2. Symbolism awareness: hand gestures, religious imagery, cultural icons with different meanings. 3. Communication styles: direct (German) vs. indirect (Japanese), high-context vs. low-context cultures. 4. Social norms: family structures, gender roles, consumer behavior patterns. Adaptation strategies: 1. Transcreation: adapt message for cultural relevance while maintaining intent. 2. Visual modifications: model diversity, clothing appropriateness, lifestyle representation. 3. Timing considerations: religious holidays, cultural celebrations, business calendars. 4. Platform preferences: WeChat (China), LINE (Japan), WhatsApp (Latin America). Local validation process: 1. Cultural consultants: native speakers with marketing expertise. 2. Focus groups: target audience feedback on cultural appropriateness. 3. Legal review: advertising regulations, claim substantiation requirements. Brand consistency: core message unchanged, execution adapted for cultural relevance. Documentation: cultural guidelines database for future campaigns, decision rationale for adaptation choices.

Traditional Italian pasta dough technique. Formula: 100g 00 flour per 1 large egg (golden ratio). Steps: 1. Create flour well on work surface. 2. Add eggs and slowly incorporate flour from inner walls. 3. Knead 10 minutes until smooth and elastic. 4. Rest 30 minutes wrapped in plastic. 5. Roll through pasta machine from setting 1 to 7. 6. Cut into desired shapes (tagliatelle, fettuccine, pappardelle). Explain gluten development, proper hydration, and how to test dough readiness.

Build viral referral programs. Mechanics: 1. Dual-sided incentives (referrer + referee). 2. Easy sharing mechanisms. 3. Trackable referral links. 4. In-product prompts at key moments. 5. Email campaigns to encourage sharing. 6. Gamification with tiers or leaderboards. 7. Social proof of referral success. 8. Fraud detection. Use tools like Viral Loops or ReferralCandy. Make rewarding feel instant.

Calculate and optimize CAC. Formula: Total Sales & Marketing Costs ÷ Number of New Customers. Best practices: 1. Segment by channel. 2. Include all costs (tools, salaries, ads). 3. Track cohorts over time. 4. Compare to LTV (LTV:CAC ratio should be 3:1). 5. Payback period (ideally < 12 months). 6. Optimize high-CAC channels. 7. Increase conversion rates. 8. Retention reduces effective CAC.

Design robust CI/CD pipelines that automate software delivery with quality gates and rollback mechanisms. Pipeline stages: 1. Source control integration: GitHub/GitLab webhooks trigger builds on commits. 2. Build automation: compile code, dependency resolution, artifact generation. 3. Testing suite: unit tests (>80% coverage), integration tests, security scans. 4. Quality gates: SonarQube analysis, vulnerability scanning, performance benchmarks. 5. Deployment stages: dev → staging → production with approval workflows. Jenkins pipeline configuration: declarative Jenkinsfile with parallel stages, environment-specific variables, credential management. GitLab CI/CD: .gitlab-ci.yml with stages, artifacts, deployment environments, manual approvals. GitHub Actions: workflow triggers, matrix builds, environment secrets, deployment strategies. Quality metrics: build success rate (>95%), deployment frequency (daily for mature teams), lead time (<1 hour for hotfixes), mean time to recovery (<30 minutes). Rollback strategies: blue-green deployments, database migration rollbacks, feature flags for instant disabling. Security integration: SAST/DAST scanning, dependency vulnerability checks, secret detection, compliance verification.

Implement safe feature releases using feature flags. Flag types: 1. Release flags: control feature deployment (temporary). 2. Experiment flags: A/B testing (temporary). 3. Ops flags: circuit breakers for performance (permanent). 4. Permission flags: user role access (permanent). Rollout strategy: 1. Internal team (0.1% traffic): validate basic functionality. 2. Beta users (1% traffic): gather feedback from friendly customers. 3. Gradual rollout (5%, 25%, 50%, 100%): monitor metrics at each stage. 4. Success criteria: error rates <0.1%, performance impact <10ms, user feedback positive. Monitoring: set up alerts for error spikes, performance regression, customer complaints. Rollback plan: instant flag toggle if issues detected. Tools: LaunchDarkly, Split, Unleash, or custom solution. Flag hygiene: remove old flags after full rollout, document flag purpose and owner.

Improve conversions with CRO. Testing framework: 1. Analyze user behavior (heatmaps, recordings). 2. Identify friction points. 3. Form hypothesis for improvement. 4. A/B test one variable at a time. 5. Statistical significance before conclusions. 6. Test headlines, CTAs, images, layouts. 7. Mobile vs desktop optimization. 8. Continuous iteration cycle. Use tools like Optimizely or VWO. Prioritize high-traffic pages.

Implement robust data governance for user privacy compliance. Data classification: 1. Public: can be shared freely (marketing content). 2. Internal: company confidential information. 3. Personal: user-identifiable information (PII). 4. Sensitive: payment data, health records, requiring encryption. Privacy compliance framework: 1. Data minimization: collect only necessary information. 2. Purpose limitation: use data only for stated purposes. 3. Consent management: clear opt-in/opt-out mechanisms. 4. Right to erasure: ability to delete user data. 5. Data portability: export user data on request. Technical implementation: 1. Encryption at rest and in transit. 2. Access controls: role-based permissions. 3. Audit logging: track data access and modifications. 4. Anonymization: remove identifiers for analytics. 5. Retention policies: automatic deletion of old data. Tools: OneTrust for consent management, Privacera for data discovery. Regular audits: quarterly privacy impact assessments, annual security reviews.

Productize consulting services. Strategy: 1. Identify repeatable deliverables. 2. Package into fixed-scope offerings. 3. Value-based pricing not hourly. 4. Templates and frameworks. 5. Tier offerings (good/better/best). 6. Clear process and timeline. 7. Reduce customization. 8. Scale with junior talent. Benefits: predictability, scalability, higher margins. Combine with retainers. Build IP assets. Move from time-for-money to leverage.

Outline a K-12 digital citizenship curriculum. Key Themes (by grade band): K-2 (Safety & Balance): online safety basics, screen time balance. 3-5 (Privacy & Communication): personal information, cyberbullying awareness, respectful online talk. 6-8 (Media Literacy & Digital Footprint): identifying fake news, understanding digital permanence, online reputation. 9-12 (Copyright & Activism): fair use, intellectual property, using social media for social good. Implementation: monthly lessons delivered by homeroom teachers, integrated into subject areas (e.g., citing sources in history), parent workshops, student-led campaigns. Use resources from Common Sense Media. Assess via scenarios and reflections.

Translate complex scientific concepts for general audiences while maintaining accuracy. Science communication principles: 1. Plain language: replace jargon with everyday terms or explain technical vocabulary. 2. Analogy and metaphor: relate complex concepts to familiar experiences. 3. Narrative structure: story arc with conflict, resolution, implications. 4. Visual support: diagrams, infographics, charts for complex data. Audience analysis: 1. Education level: high school to college graduates for general public. 2. Prior knowledge: assume basic science literacy, build from there. 3. Interests: health, environment, technology applications. 4. Reading preferences: scanning vs. deep reading, mobile vs. desktop. Accuracy considerations: 1. Source verification: peer-reviewed research, expert interviews. 2. Uncertainty acknowledgment: discuss limitations, ongoing research. 3. Context provision: broader implications, real-world applications. 4. Expert review: scientist fact-checking before publication. Writing techniques: 1. Inverted pyramid: key findings first, methodology details later. 2. Active voice: makes writing more engaging and clear. 3. Concrete examples: specific cases rather than abstract concepts. Distribution channels: science blogs, popular magazines, social media threads, podcast scripts.

Implement charts quickly with Chart.js. Chart types: 1. Line charts for trends. 2. Bar charts for comparisons. 3. Pie/Doughnut for proportions. 4. Radar for multi-axis data. 5. Scatter for correlations. 6. Mixed chart types. 7. Responsive and accessible. 8. Plugin system for customization. Use react-chartjs-2 wrapper and implement real-time updating with data push.

Optimize push notifications for engagement. Strategy: 1. Permission priming before asking. 2. Segment users for relevance. 3. Personalize based on behavior. 4. Optimal timing based on user activity. 5. Clear, actionable messages. 6. Rich notifications with images. 7. Deep linking to specific content. 8. A/B test copy and timing. Monitor opt-out rates. Use OneSignal or Firebase. Don't overuse.

Build comprehensive DevOps metrics dashboards for measuring team performance and continuous improvement initiatives. DORA metrics (DevOps Research and Assessment): 1. Deployment frequency: daily deployments for elite teams, weekly for high performers. 2. Lead time for changes: <1 hour for elite teams, <1 week for high performers. 3. Mean time to recovery (MTTR): <1 hour for elite teams, <1 day for high performers. 4. Change failure rate: 0-15% for elite teams, 16-30% for high performers. Pipeline metrics: 1. Build success rate: >95% target, trend analysis, root cause analysis for failures. 2. Test coverage: >80% code coverage, test execution time, test reliability metrics. 3. Security scanning: vulnerability detection rate, time to remediation, policy compliance. 4. Infrastructure metrics: provisioning time, resource utilization, cost per deployment. Quality metrics: 1. Code quality: technical debt ratio, code duplication, maintainability index. 2. Bug escape rate: production bugs vs. bugs found in testing, customer-reported issues. 3. Performance: response time trends, error rate tracking, SLA compliance. Business alignment: 1. Feature delivery: story points delivered, cycle time, value delivered to customers. 2. Customer satisfaction: NPS scores, support ticket volume, feature adoption rates. Dashboard tools: 1. Grafana: metric visualization, alerting, data source integration. 2. Datadog: APM integration, real-time monitoring, anomaly detection. 3. Splunk: log analysis, ITSI for service insights, business KPI correlation. Automation: scheduled reports, alert thresholds, trend analysis, predictive analytics for capacity planning.

Create business model with canvas. Components: 1. Value Propositions (what you offer). 2. Customer Segments (who you serve). 3. Channels (how customers find you). 4. Customer Relationships (engagement type). 5. Revenue Streams (how you make money). 6. Key Resources (assets needed). 7. Key Activities (what you do). 8. Key Partnerships (who helps). 9. Cost Structure (expenses). Use for lean validation and iteration.

Engage multiple stakeholders in target accounts. Multi-threading definition: relationships with 3+ people in buying org. Why: single-threaded deals stall when champion leaves or loses political battle. Strategy: 1. Map org chart (LinkedIn, ZoomInfo). 2. Identify 5-7 key stakeholders. 3. Assign custom approach per person. Example: Economic Buyer (exec briefing, ROI focus), Champions (detailed demos, frequent touch), End Users (hands-on trial, training resources). Tactics: ask champion for intros ('Who else should evaluate this?'). Attend prospect events/conferences. Engage on LinkedIn. Send personalized gifts. Track relationship depth (0=unaware, 1=aware, 2=engaged, 3=advocate). Safeguard deals: if 1 person goes dark, others keep deal alive.

Write professional screenplays using industry-standard formatting and structure. Screenplay format: 1. Courier 12-point font, 1 page = 1 minute screen time. 2. Scene headings: INT./EXT. LOCATION - TIME OF DAY. 3. Action lines: present tense, active voice, visual descriptions only. 4. Character names: CENTERED, CAPS when introduced and before dialogue. 5. Dialogue: centered, natural speech patterns, subtext over exposition. Three-act structure timing: Act I (25 pages): setup, character introduction, inciting incident at page 12-15. Act II (50 pages): rising action, midpoint at page 50-60, obstacles escalate. Act III (25 pages): climax, resolution, denouement. Plot points: 1. Inciting incident: event that starts story in motion. 2. Plot point I (end Act I): major story turn. 3. Midpoint: game-changing revelation or setback. 4. Plot point II (end Act II): final obstacle before climax. Character development: protagonist arc, antagonist motivation, supporting character functions. Industry guidelines: 90-120 pages for features, character names in CAPS only first appearance, minimal camera directions. Software: Final Draft (industry standard), WriterDuet (collaboration), Celtx (free option).

Develop a custom Vite 5 plugin for specialized workflow. Plugin features: 1. Transform hook for custom file types. 2. Virtual modules for config injection. 3. HMR API for hot updates. 4. Dev server middleware. 5. Build hooks for asset processing. 6. SSR transform handling. 7. Dependency pre-bundling optimization. 8. Watch mode for external tools. Use TypeScript with proper plugin typing and implement configureServer for custom routes.

Practice test-driven development. Workflow: 1. Write failing test first (Red). 2. Write minimal code to pass (Green). 3. Refactor while keeping tests green. 4. Repeat cycle. Benefits: Better design, confidence, documentation. Write tests for: edge cases, error handling, happy path. Use describe/it structure. Keep tests fast and isolated. Mock external dependencies.