David RylandSenior Software Engineer • AI Engineer
Summary

Senior Software Engineer and AI Engineer with 10+ years of experience designing, building, and scaling production distributed systems, cloud-native applications, and AI-powered platforms. Expertise in Python, FastAPI, Go, TypeScript, microservices, asynchronous programming, scalable API architecture, and high-performance backend services. Deep experience in Generative AI, retrieval-augmented generation (RAG), agentic workflows, LLM orchestration, vector search, multimodal retrieval, and production AI evaluation. Delivered retrieval infrastructure supporting 20M+ users and 10K+ QPS with 10–100 ms P99 latency, real-time conversational AI platforms, and healthcare systems supporting 5,000+ concurrent users. Proven ownership across system design, full-stack delivery, cloud infrastructure, performance optimization, observability, CI/CD, and cross-functional technical leadership.

Core Skills
Languages & Backend: Python, Go, TypeScript, JavaScript, Java, C#, FastAPI, Django, Node.js, NestJS, Spring Boot, .NET Core, REST APIs, GraphQL, WebSockets, Asynchronous Programming|Frontend & Real-Time Systems: React, Next.js, Angular, Redux, LiveKit, WebRTC, Twilio, Whisper, Deepgram, ElevenLabs, FFmpeg, Real-Time Audio Streaming, Speech-to-Text, Text-to-Speech|Software Architecture & Distributed Systems: System Design, Distributed Systems, Microservices, Event-Driven Architecture, Concurrent Processing, High-Throughput Services, Low-Latency Systems, Scalable APIs, State Management, Service Integration, Performance Optimization, Redis Caching, Cloud-Native Architecture|Generative AI & Agentic Systems: Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), LangGraph, LlamaIndex, Multi-Agent Orchestration, Agentic Workflows, Tool/Function Calling, Prompt Engineering, Human-in-the-Loop, Context Management, Structured Retrieval, Conversational AI, Gemini, Vertex AI|Retrieval, Databases & Data Engineering: PostgreSQL, Redis, BigQuery, FAISS, pgvector, Vector Search, Hybrid Search, Semantic Ranking, Embeddings, Approximate Nearest Neighbor (ANN), Query Rewriting, Multimodal and Multilingual Retrieval, ETL, Batch Processing, Databricks, Delta Lake|AI Evaluation, Inference & Observability: LangSmith, Langfuse, Retrieval Evaluation, Prompt Evaluation, Offline Model Scoring, Execution Tracing, Model Drift Monitoring, Hallucination Monitoring, Model Serving, ONNX, Quantization, Triton Inference Server, GPU/CPU Optimization, Latency Analysis|Cloud, DevOps & Testing: AWS (EC2, Lambda, S3, RDS), GCP (GKE, Cloud Run, Cloud Functions, Pub/Sub, Vertex AI, BigQuery, Cloud Storage), Docker, Kubernetes, Terraform, Infrastructure as Code, GitHub Actions, CI/CD, Automated Testing
Professional Experience
Striveworks LLC, Senior FullStack Engineer / AI Engineer
01/2025 – 10/2026 | Remote
  • •Owned the architecture and production delivery of a real-time AI voice agent platform, designing Python/FastAPI and Django backend services integrated with LiveKit, WebRTC, Twilio, and GCP infrastructure to support low-latency conversational workflows and enterprise integrations.
  • •Designed a distributed, event-driven processing architecture using GCP Cloud Run, Cloud Functions, Pub/Sub, and Cloud Storage, separating streaming requests, asynchronous processing, inference workloads, and data ingestion into scalable service components.
  • •Engineered high-performance WebSocket APIs and bidirectional audio-streaming services for concurrent conversational sessions, coordinating speech recognition, LLM inference, audio synthesis, and agent state while managing strict latency requirements.
  • •Built real-time speech-to-speech AI pipelines using Whisper, Deepgram, ElevenLabs, and Gemini, achieving sub-second transcription and response generation while integrating streaming inference into interactive applications.
  • •Architected production multimodal RAG pipelines using Gemini 2.5 Pro, Vertex AI Embeddings, FAISS, and pgvector, integrating audio, text, and metadata retrieval with contextual grounding to improve factual relevance and reduce unsupported LLM responses.
  • •Developed agentic workflows using LangGraph and LlamaIndex, orchestrating multi-step reasoning, tool/function calling, contextual memory, and stateful execution with human-in-the-loop checkpoints and failure-aware workflow handling.
  • •Designed event-driven conversational state machines for turn-taking, interruption detection, barge-in handling, and context preservation, improving real-time interaction behavior across speech-processing and inference components.
  • •Built ingestion and preprocessing services using FFmpeg, audio segmentation, metadata extraction, and input validation to transform unstructured audio into reliable inputs for downstream retrieval, inference, and evaluation workloads.
  • •Implemented AI evaluation and observability workflows using batch inference, dataset generation, offline scoring, and model-output validation to assess speech processing, retrieval quality, and conversational behavior and support iterative model improvements.
  • •Improved engineering delivery through Docker-based containerization, automated testing, GitHub Actions, and CI/CD pipelines, supporting repeatable application deployments and maintainable backend services.
  • •Led cross-functional implementation of production AI features, partnering with clinical stakeholders, speech researchers, and product teams on technical requirements, architecture tradeoffs, quality validation, and troubleshooting; mentored junior engineers through code and architecture reviews.
  • Zillow, Senior FullStack Engineer / AI Engineer
    11/2022 – 12/2024 | Seattle WA
  • •Architected and productionized large-scale multimodal and multilingual retrieval infrastructure serving 20M+ users and supporting 10K+ QPS with 10–100 ms P99 latency, using GCP, GKE, Cloud Run, BigQuery, and Vertex AI.
  • •Designed high-throughput Python/FastAPI and Django microservices with Node.js, Redis, and PostgreSQL integrations, exposing scalable retrieval and AI capabilities through backend APIs consumed by customer-facing applications.
  • •Optimized vector retrieval infrastructure using FAISS approximate nearest neighbor indexing and Redis-backed serving, improving retrieval efficiency, search recall, and query latency for high-volume workloads.
  • •Architected conversational search and RAG applications combining query rewriting, structured attribute extraction, embedding retrieval, semantic ranking, and contextual grounding, connected to React, Next.js, and TypeScript user interfaces.
  • •Designed distributed data-generation and model-training workflows using cross-lingual contrastive learning, synthetic datasets, and AWS GPU infrastructure, expanding multilingual embedding coverage and model-development throughput.
  • •Developed multi-step AI retrieval workflows using LangGraph and LlamaIndex, integrating automated query processing, tool/function calling, state management, and agent orchestration into search applications.
  • •Built production AI observability and governance capabilities using LangSmith, Langfuse, PostgreSQL, and custom telemetry dashboards to monitor prompt behavior, retrieval quality, hallucinations, model drift, execution traces, and auditability.
  • •Improved model-serving efficiency through ONNX deployment, quantization, Triton Inference Server integration, GPU/CPU tuning, and inference latency analysis, balancing retrieval performance with infrastructure efficiency.
  • •Developed offline evaluation and batch-processing pipelines for retrieval scoring, synthetic data generation, threshold tuning, and model-quality analysis, using BigQuery analytics and Next.js dashboards to support engineering decisions.
  • •Integrated multimodal AI prototypes combining Gemini Pro Vision, speech recognition, embeddings, and media-based retrieval to explore additional natural-language and visual search experiences.
  • •Led end-to-end technical delivery of AI search and embedding services, coordinating with research, product, and engineering teams on architecture, implementation, service integration, performance optimization, and production rollout.
  • Banner Health, FullStack Developer
    08/2019 – 10/2022 | Phoenix, AZ
  • •Designed and delivered clinical decision-support applications using Python, FastAPI, React, Angular, and TypeScript, building integrated full-stack systems for clinical search, treatment guidance, and access to patient information.
  • •Built a production hybrid retrieval platform across 1M+ clinical documents, integrating vector retrieval, semantic ranking, and contextual grounding to achieve a 22% improvement in retrieval accuracy.
  • •Architected and optimized scalable backend services using Python, FastAPI, Go, and asynchronous processing, supporting 5,000+ concurrent users while reducing application latency by 35%.
  • •Developed REST and streaming APIs for healthcare applications, integrating service authentication, clinical data access, database operations, and backend workflows into maintainable service architectures.
  • •Engineered document-intelligence and workflow-automation services for clinical summarization, information extraction, and compliance-oriented processing, reducing repetitive manual work and improving document accessibility.
  • •Designed ingestion and validation pipelines for unstructured clinical documents and structured healthcare data, supporting extraction, metadata processing, and downstream search and automation workflows.
  • •Built accessible clinician-facing web applications using React, Angular, and TypeScript, integrating reusable frontend components with clinical workflows and backend services.
  • •Improved software delivery through Docker, GitHub Actions, automated testing, and CI/CD pipelines while collaborating with clinicians and operations teams on requirements, system validation, and production integration.
  • GoDaddy, Software Developer
    07/2016 – 07/2019 | Tempe, AZ
  • •Designed and maintained production backend services using Node.js, NestJS, Java/Spring Boot, and .NET Core, developing scalable APIs, service integrations, and business application functionality.
  • •Built REST and GraphQL APIs with authentication, database connectivity, and reusable service interfaces, improving consistency and reliability across frontend and backend integrations.
  • •Delivered full-stack web applications using React, Angular, Next.js, TypeScript, and Redux, implementing responsive user interfaces and maintainable component-based architectures.
  • •Developed shared TypeScript and Storybook component libraries, improving frontend consistency, code reuse, and development efficiency across applications.
  • •Implemented AWS infrastructure and deployment automation using Terraform, Lambda, and CI/CD workflows, strengthening cloud deployment consistency and application reliability.
  • •Engineered automated media-processing and ETL-style workflows using AWS Lambda, FFmpeg, and Delta Lake, supporting batch processing and video transformation workloads.
  • •Optimized Node.js backend services and AWS compute resources to improve application performance, operational reliability, and infrastructure cost efficiency.
  • Education
    Bachelor of Science in Computer Science, Arizona State University
    08/2012 – 05/2016 | Tempe, AZ