My background is a near-exact match for what Wisedocs needs. A few highlights: *ML Systems at Scale in Healthcare* I currently architect PACE (Patient Access Care Engine) at IKS Health — a LangGraph-based platform with Gemini 2.5 Pro orchestrating 4 specialized subagents, 32 tools, and 10 deployed graphs for autonomous patient engagement across 300K+ patients. I've built the full pipeline: Elasticsearch patient search across 12+ query dimensions, Firestore persistence, Cloud Tasks async scheduling, and real-time event streaming. Scaling ML inference in healthcare is literally what I do every day. *LLMs, RAG & BERT in Production* At Adaapt AI, I built an agentic RAG system with multi-step reasoning and multi-LLM cascade (Claude/OpenRouter/OpenAI) achieving 99.9% uptime and sub-500ms responses. I also built a document intelligence platform supporting 65+ file formats with 97.9% table extraction accuracy — directly relevant to Wisedocs' medical document parsing challenge. *Scalable Python APIs & Infrastructure* Every role I've held has involved building production FastAPI services with Docker/Kubernetes, async processing (Celery/Redis), and cloud deployment (AWS ECS, GCP Cloud Run). At Outspeed, I optimized multimodal model inference to achieve 1.75x speedup and 79% memory reduction — the kind of work needed to handle 50K-page PDFs at 100x scale. *Cross-Team Collaboration* As Head of Engineering at General Magic, I led a team through 5 major releases with 30% faster feature delivery, mentoring engineers and establishing AI best practices. I'm comfortable working across ML, backend, and product teams to ship. *Why Wisedocs specifically:* I've spent the last year deep in healthcare AI — building systems that process medical records, manage patient campaigns, and integrate with clinical workflows. The intersection of ML + healthcare documents + scale is exactly where my expertise sits.…My background is a near-exact match for what Wisedocs needs. A few highlights: *ML Systems at Scale in Healthcare* I currently architect PACE (Patient Access Care Engine) at IKS Health — a LangGraph-based platform with Gemini 2.5 Pro orchestrating 4 specialized subagents, 32 tools, and 10 deployed graphs for autonomous patient engagement across 300K+ patients. I've built the full pipeline: Elasticsearch patient search across 12+ query dimensions, Firestore persistence, Cloud Tasks async scheduling, and real-time event streaming. Scaling ML inference in healthcare is literally what I do every day. *LLMs, RAG & BERT in Production* At Adaapt AI, I built an agentic RAG system with multi-step reasoning and multi-LLM cascade (Claude/OpenRouter/OpenAI) achieving 99.9% uptime and sub-500ms responses. I also built a document intelligence platform supporting 65+ file formats with 97.9% table extraction accuracy — directly relevant to Wisedocs' medical document parsing challenge. *Scalable Python APIs & Infrastructure* Every role I've held has involved building production FastAPI services with Docker/Kubernetes, async processing (Celery/Redis), and cloud deployment (AWS ECS, GCP Cloud Run). At Outspeed, I optimized multimodal model inference to achieve 1.75x speedup and 79% memory reduction — the kind of work needed to handle 50K-page PDFs at 100x scale. *Cross-Team Collaboration* As Head of Engineering at General Magic, I led a team through 5 major releases with 30% faster feature delivery, mentoring engineers and establishing AI best practices. I'm comfortable working across ML, backend, and product teams to ship. *Why Wisedocs specifically:* I've spent the last year deep in healthcare AI — building systems that process medical records, manage patient campaigns, and integrate with clinical workflows. The intersection of ML + healthcare documents + scale is exactly where my expertise sits.WWWW…