Emilio Esposito

    Senior Director, AI Engineering & Enablement at LegalZoom
    Co-founder & Managing Partner, Sernia Capital

    Emilio Esposito

    I lead AI Engineering & Enablement at LegalZoom, combining engineering leadership with hands-on software development. I built the function from the ground up and still write production code daily, building AI products and helping a 200-person engineering org put AI tools to practical use.

    I also co-founded and operate Sernia Capital, a 40-unit residential real estate portfolio. Running the business gives me plenty of reasons to build software. This site brings together some of that work, a few experiments, and tools we use day to day.

    GitHubLinkedIn

    Software for Sernia Capital

    A few systems I've built to help run our properties, from urgent tenant messages to leasing follow-ups. Most are part of our daily operations; others are still in development.

    Sernia AI: Operations Assistant

    Production

    An assistant for property managers, available through Quo SMS or our web app. It uses business communications and persistent memory to keep context, manage tasks, and help with follow-ups, with approval required for actions like sending external messages. It runs from chats, incoming events, or scheduled check-ins. Built with PydanticAI.

    Emergency SMS Routing

    Production

    AI checks incoming tenant messages for urgent issues and escalates them through Twilio calls and texts, including off-hours alerts configured to bypass Do Not Disturb.

    View source

    Leasing Lead Management

    Production

    Turns Zillow email threads into contacts, calendar events, and follow-up reminders for leasing agents, with contact details synced to our phone platform.

    View source

    Tenant Communications

    Production

    Building-wide announcements by SMS, with role-based access so property managers can reach the right tenants.

    Open app

    AI Leasing Auto-Replies

    In Development

    Drafting and testing automated replies to leasing inquiries using property details, listings, and agent availability, with applicant screening before suggesting a showing.

    Preview

    Tools & workflow

    What I reach for to build, ship, and maintain software.

    Development

    Codex

    My first choice since June 2026 for day-to-day development, from exploring a codebase to implementing features and reviewing changes.

    Claude Code

    A very close second, used in parallel with Codex. Still a regular part of my workflow for features and debugging. I also led its rollout across 200 engineers at LegalZoom.

    Parallel worktrees

    A first-class part of every project. Isolated Git worktrees let AI agents tackle separate tasks at the same time, each with its own branch and development environment.

    Cloud maintenance agents

    Cloud-based AI agents investigate issues and handle routine software maintenance, triggered by Logfire alerts or scheduled runs.

    AI & agents

    OpenRouter

    My API inference layer for frontier and open-source models. It makes it easy to choose the right model for each job and balance quality with cost, with governance and guardrails for agent spending, model access, and data privacy.

    PydanticAI

    My usual starting point for Python agents: structured outputs, dependency injection, tool calling, and graphs for workflows that need multiple agents.

    FastMCP

    For exposing tools, resources, and prompts through the Model Context Protocol. I use it for MCP servers at LegalZoom and in my own projects.

    Frameworks

    FastAPI

    Python APIs with async support, Pydantic validation, and generated documentation. A straightforward fit for the backends I build.

    React Router v7

    Loaders, actions, and nested routes keep data fetching close to the UI. Vite keeps the local development loop fast.

    Infrastructure

    Railway

    Hosting for my apps and services, with monorepo support and PR preview environments. The CLI and MCP tools fit into my development workflow.

    Logfire

    Traces and logs for Python and AI workloads, including PydanticAI. I use its MCP server to investigate production issues and its alerts to trigger maintenance agents.

    Neon Postgres

    Postgres with database branching. Each PR preview gets an isolated database, with setup and cleanup automated.

    Selected open source

    A few things I've built and shared.

    Agent Filetree Memory MCP

    Early alpha

    Persistent memory for AI agents as a Markdown file tree, backed by PostgreSQL with encryption at rest and version history. Use it through MCP or embed it in a Python service.

    This site & Sernia tools

    Source for this site and the Sernia systems featured above: a React Router frontend, a FastAPI backend, and the integrations that connect our day-to-day operations.