Alexander Weaver Senior Software Engineer
Summary

Senior backend and platform engineer with deep expertise in AI-driven distributed systems, large-scale data pipelines, real-time runtimes, and high‑performance backend architecture. I design and build scalable, resilient, and secure systems that power AI agents, multimodal pipelines, and mission‑critical analytics workloads. My experience spans LLM orchestration, microVM execution environments, REST API design, cloud infrastructure, and data processing systems across healthcare, real estate, commerce, and enterprise environments. I thrive in early-stage, high‑ownership roles where backend engineering directly shapes product direction and customer impact.

Skills
Programming Languages

JavaScript, TypeScript, Python, Go, Java

Backend & Platform Engineering

Node.js, Go, Python (FastAPI), Elixir, Phoenix, Absinthe, REST APIs, GraphQL, service-oriented architecture, backend scalability, performance optimization, HIPAA-compliant backend systems, RBAC, encryption, audit logging, serverless runtimes, microVM execution environments

AI / Machine Learning & GenAI Systems

OpenAI, Claude, Gemini, Gemini Enterprise, DeepSeek, LangChain, LangGraph, LangSmith, RAG pipelines, AI agents, multimodal systems (STT → LLM → TTS), real-time conversational AI, hybrid search, AI observability

Cloud, DevOps & Reliability Engineering

AWS, GCP, Azure, Docker, Kubernetes, Terraform, Vercel, GitHub Actions, CircleCI, Jenkins, OpenTelemetry, Prometheus, logging & monitoring systems, serverless compute (Lambda), microVM orchestration, Vertex AI Pipelines, Vertex AI Vector Search, Vertex AI Monitoring

Frontend Architecture

React, Next.js, TypeScript, Redux, Zustand, React Query, WebRTC UI flows, component-driven architecture, Storybook, Tailwind CSS, Material UI, SCSS, performance optimization (code splitting, lazy loading, bundle analysis), real-time browser communication (WebSockets, WebRTC)

System Design & Architecture

Distributed systems, scalable system design, event-driven architecture, microservices, real-time systems, multimodal AI pipelines, secure sandboxed runtimes (v8 isolates, Firecracker microVMs), API design, system integration, authentication & authorization, caching strategies, message-driven architectures (Kafka), high-availability systems, retrieval-grounded workflows

Data Systems

PostgreSQL, MySQL, MongoDB, DynamoDB, Redis, pgvector, Pinecone, FAISS, real-time data synchronization, distributed caching, event-driven data pipelines

Experience
MaxHome.AI, Senior Software Engineer
01/2025 – 03/2026 | Richardson, TX
  • •Engineered the architecture, infrastructure, and execution environment for MaxHome’s AI automation platform, building backend, runtime, and agent-orchestration layers with TypeScript, PostgreSQL, event-driven pipelines, and encryption; this enabled faster feature delivery and improved system reliability.
  • •Found that LLMs generate code more reliably than tool-call sequences and designed a secure, multi-layer sandbox for executing LLM-generated Python and TypeScript, reducing execution errors and strengthening code safety.
  • •Built a dual-runtime sandbox system that combines a fast serverless JavaScript runtime with a Firecracker-powered microVM, delivering low-latency execution for lightweight tasks and secure, isolated environments for complex workloads.
  • •Implemented a serverless JavaScript runtime using V8 isolates that launches isolated execution contexts in ~20 ms, providing low-latency processing for lightweight tasks.
  • •Developed a Firecracker-powered microVM environment that cold-boots a fully isolated Python/TypeScript runtime in ~2 s, supporting complex workloads while maintaining strong security isolation.
  • •Integrated the sandbox into MaxHome’s AI voice agent platform, enabling agents to run custom TypeScript tools, execute LLM-generated code, and interact with real-time property data.
  • •Integrated Gemini Enterprise models into real‑time conversational and property‑analysis pipelines, improving accuracy and reducing hallucinations.
  • •Built the WebRTC frontend for real-time conversational experiences, handling RTP timing events, streaming audio, and bidirectional communication with Twilio and browser clients.
  • •Designed and shipped a multimodal STT → LLM → TTS pipeline using Gemini Live API, OpenAI Realtime, and ElevenLabs to support natural, low-latency property-tour conversations.
  • •Developed a chat client that supports serverless tool calling, coding agents, and hybrid search to ground LLM outputs in property, zoning, and neighborhood datasets.
  • •Used Claude Code to generate safe, auditable TypeScript and Python execution modules inside the sandbox environment.
  • •Leveraged Claude Code’s structured tool‑use patterns to reduce hallucinations in code‑generation workflows and improve reliability of agent‑driven automation.
  • •Integrated Claude Code into the multimodal pipeline for deterministic code‑execution tasks, enabling agents to produce validated code patches, data‑transformation scripts, and property‑analysis utilities.
  • •Built Urai Portal, enabling MaxHome teams to run open-source local models alongside cloud LLMs, mixing and matching them within a single conversation to optimize cost, latency, and accuracy.
  • •Defined Urai Chat, an AI-powered real-estate assistant that combines code-execution agents, retrieval-augmented reasoning, and multimodal interaction, using TypeScript, event-driven data pipelines, PostgreSQL, encryption, LangGraph, and LangSmith, which streamlined property queries and improved user satisfaction.
    • OPTUM, Full Stack Engineer
    11/2022 – 12/2024 | Miami, FL
  • •Built LLM-powered helper tooling connected to internal APIs and MCP-style interfaces, enabling unified investigation across patient events, payments, CRM records, orders, third-party provider data, alerts, and distributed system failures.
  • •Shifted AI agents from basic “write code” prompts to operational diagnosis, enabling LLMs to analyze production systems, trace user-impacting issues, and provide actionable remediation steps.
  • •Designed retrieval-grounded workflows that combined clinical guidelines, historical notes, and operational telemetry, improving accuracy and reducing hallucination risk in clinician-facing assistants.
  • •Implemented pgvector-based semantic retrieval to unify search across clinical documentation, operational logs, and secure messaging systems.
  • •Added AI observability including drift detection, latency monitoring, and structured model-behavior logging to improve reliability and accelerate root-cause analysis.
  • •Integrated Gemini Enterprise models into clinician‑facing assistants for guideline retrieval, patient‑event analysis, and operational diagnostics.
  • •Built Vertex AI Workbench notebooks and automated MLOps workflows using Vertex Pipelines, Cloud Functions, and Pub/Sub.
  • •Implemented Vertex AI Vector Search for unified semantic retrieval across clinical notes, PHI‑safe logs, and operational telemetry.
  • •Architected HIPAA‑aligned Vertex AI deployments with encryption, RBAC, VPC‑SC, and PHI‑safe data flows.
  • •Built HIPAA-compliant backend services with RBAC, encryption, audit logging, and PHI-safe data flows, ensuring readiness for internal and external security audits.
  • •Collaborated with clinicians, compliance teams, and operations engineers to validate AI-assisted workflows, ensuring safe deployment in regulated healthcare environments.
    • MarqVision, Backend Developer
    05/2020 – 09/2022 | Los Angeles, CA
  • •Built high-performance storefronts and merchant dashboards using React, Next.js, TypeScript, Redux, Zustand, React Query, and server-side rendering to support thousands of concurrent shoppers.
  • •Designed and implemented backend microservices in Node.js, Elixir/Phoenix, Absinthe, and Python (FastAPI) for catalog management, pricing, promotions, checkout, and order orchestration, improving order-processing reliability and reducing latency for customer transactions.
  • •Architected real-time inventory and pricing systems using PostgreSQL, Redis, DynamoDB, Kafka, and event-driven microservices to ensure consistency across storefronts, warehouses, and third-party logistics providers.
  • •Developed GraphQL and REST APIs powering merchant tools, storefront rendering, and partner integrations, improving API performance and reducing latency across global regions.
  • •Built serverless functions on AWS Lambda for high-volume workloads including cart updates, tax calculations, and fraud-detection triggers, enabling near-real-time transaction processing and lowering compute costs.
  • •Improved checkout performance by 35% through caching strategies, query optimization, and distributed session management.
  • •Created internal SDKs in TypeScript and Python for partner developers to integrate with BigCommerce’s commerce engine, reducing integration time and improving developer adoption.
  • •Enhanced platform reliability with OpenTelemetry tracing, structured logging, and automated CI/CD pipelines using GitHub Actions and CircleCI.
  • •Collaborated with product, design, and SRE teams to launch three new merchant features and improve storefront load time, maintaining 99.9% uptime during peak retail events.
    • Google, Frontend Developer
    04/2018 – 04/2020 | Lubbock, TX
  • •Built high-performance, responsive web applications using React, Next.js, TypeScript, and modern JavaScript, improving load times, interactivity, and overall user experience across multiple Google product surfaces.
  • •Collaborated with UX/UI designers, product managers, and accessibility teams using Figma and Google’s internal design system to deliver intuitive interfaces that met performance and accessibility standards, resulting in higher user satisfaction.
  • •Optimized frontend performance through code splitting, lazy loading, tree-shaking, bundle analysis, and build-level optimizations, reducing initial load times and improving Lighthouse scores.
  • •Integrated frontend applications with backend services using REST APIs, GraphQL, and WebSockets, enabling seamless real-time data flow and interactive user experiences.
  • •Implemented component-driven architecture using reusable UI patterns, Storybook, and internal Google tooling to improve consistency and velocity.
  • •Strengthened application reliability through unit testing, integration testing, and end-to-end testing using Jest, Cypress, and internal Google test frameworks.
  • •Improved developer workflows by contributing to internal libraries, shared UI components, and build tooling, reducing onboarding time and improving cross-team collaboration.
    • Google, Software Engineering Intern
    01/2018 – 03/2018 | Lubbock, TX
  • •Supported development of internal full-stack tools with Python (FastAPI), Java, and JavaScript, adding new API endpoints and UI features that enabled engineers to automate routine tasks.
  • •Assisted in building backend components on GCP using Cloud Functions and Pub/Sub, creating data-processing workflows and internal APIs that reduced manual data handling for operations teams.
  • •Developed React and TypeScript UI components for internal dashboards, fixing layout bugs and improving usability, which shortened the time users spent navigating the dashboards.
  • •Performed testing and debugging with pytest and conducted code reviews in GitHub, catching defects early and maintaining compliance with Google engineering standards.
  • •Learned scalable system design and distributed architecture principles through mentorship, applying them to design a real-time WebSocket service prototype that demonstrated feasibility for future projects.
  • •Improved developer workflows by updating internal documentation, refining build scripts, and contributing to shared libraries used across multiple teams.
    • Education
    Bachelor's, Computer Science, Texas Tech University
    01/2013 – 12/2017 | Lubbock, TX
    Certificates
    Google Advanced Data Analytics Certificate
    Google AI / Generative AI Skill Badges
    IBM Generative AI Engineering Certificate