David Lapsley, Ph.D.
CTO, ActualyzeAI
Welcome!
Hi, I’m Dave! With 25+ years building infrastructure platforms at AWS, Cisco, and now as CTO of ActualyzeAI, I help enterprises navigate the journey from AI proof-of-concept to production deployment. This blog shares battle-tested patterns, real-world case studies, and practical strategies for the 13% of AI projects that actually make it to production.
Why 87% of AI projects fail isn’t a technical problem—it’s a planning problem. Let’s fix that.
Current Projects
NVIDIA DGX Spark Hardware Research
Comparing NVIDIA’s integrated AI platform ($3,999) against DIY open-source GPU infrastructure. 8-week testing series covering setup time, performance, cost, and developer experience.
Topics: Hardware comparison, model serving, local AI infrastructure
Tampa Bay Enterprise AI Community
Monthly meetups bringing together CTOs, platform engineers, and business leaders to discuss production AI infrastructure challenges.
Next Event: November 14, 2025 - “Hands-On: Building Production AI Infrastructure with KServe”
Join the community:
Featured Topics
Enterprise AI Infrastructure
Production deployment patterns, Kubernetes for AI workloads, GPU cost optimization, hybrid cloud architectures, and infrastructure that scales from POC to production.
AI Governance & Compliance
HIPAA, SOX, GDPR compliance for AI systems. Data sovereignty, model governance, audit trails, and risk management frameworks for regulated industries.
Cost Optimization & ROI
GPU utilization strategies, spot instance management, resource allocation, chargeback models, and TCO analysis for cloud vs. on-premises AI infrastructure.
MLOps & Platform Engineering
KServe, KubeFlow, model serving, CI/CD for ML, monitoring, multi-tenancy patterns, and building production-grade AI platforms.
Upcoming Content
November 2025: Hands-On KServe Workshop
Step-by-step guide to deploying production AI infrastructure with KServe in 60 minutes. Complete deployment automation scripts and Kubernetes manifests included.
Topics covered:
- Kubernetes cluster + GPU Operator setup
- Istio + Knative installation
- KServe deployment and configuration
- GPU-accelerated InferenceService deployment
- Monitoring with Prometheus, Grafana, and DCGM