Alex Beattie - Senior AI/LLM Engineer
Senior AI/LLM engineer and founding engineer focused on production AI systems, healthcare software, and scalable Django/Python backends.
I work across the parts of AI engineering that determine whether a system survives production: request routing, grounding, source metadata, streaming contracts, eval harnesses, observability, model routing, and cloud cost controls.
Core Focus
- Production AI systems: routing, tool selection, safety gates, Strands agent behavior, and observable decision flows
- RAG and retrieval: pgvector, Pinecone, Cohere embeddings, source metadata, query rewriting, grounding, and retrieval evaluation
- LLM evaluation: Langfuse traces, gold-case evals, LLM-as-judge workflows, deterministic retrieval metrics, and live trace scoring
- Model economics: cost-aware routing, token analytics, budget alerts, and smaller-model paths for simple queries
- Backend architecture: Django, Python, Strawberry GraphQL, PostgreSQL, Redis, and API contracts
- Cloud infrastructure: AWS Bedrock, ECS, Aurora, S3, GitHub Actions, and production cost optimization
- Mobile-connected systems: Flutter clients, streaming APIs, auth flows, and mobile/backend contracts
- AI development workflows: Cursor custom rules/skills, Claude Code, MCP servers, and IDE-native engineering context
Selected Writing
- Designing Production AI Routing And Evals For A Healthcare Assistant
- Lessons From Bedrock, pgvector, And RAG In Production
- Streaming LLM Architecture Patterns: Sources, Done Events, And Observability
Selected Projects
- Stella / Ourself Health - production women’s health AI assistant work at Ourself Health, including Django/GraphQL backend architecture, AWS Bedrock, Strands agents, pgvector RAG, Langfuse observability/evals, streaming responses, and mobile/backend contracts.
- KiNDD / NDD Resource Navigator - founder/engineer nonprofit startup platform with a live web map, iOS app, Django/PostGIS backend, Bedrock AI assistance, and Cohere/Pinecone research RAG.
- Unified Connectors MCP Server - MCP-first engineering workflow layer for GitHub, Slack, Jira, Confluence, Google Docs, and Miro.
KiNDD writing:
- Building A Geospatial Care-Navigation Platform For Developmental Disability Services
- Designing AI Assistance For A Public-Interest Healthcare Navigator
- From Commissioned Website To Nonprofit Startup: The KiNDD Technical Journey
Experience
I have 14+ years of experience shipping software across mobile, backend, cloud, and AI systems. In the last year, my work has centered on two production AI products: Stella at Ourself Health and KiNDD / NDD Resource Navigator. My work has moved from Swift/iOS and TypeScript/Node.js into Python-first backend and AI infrastructure, with a current focus on healthcare AI systems that need reliable retrieval, evaluation, observability, and operational discipline.
Before healthcare AI, I spent 13 years as the sole technologist for a real estate firm, owning software, automation, web systems, integrations, and technical operations end to end.