Code BEAM America 2025
who it's for
Elixir developers, Erlang developers, BEAM engineers, backend engineers, embedded engineers, site reliability engineers, platform engineers
topics
Phoenix LiveView with Web Components; LiveElements integration patterns; Nerves Linux with Flutter; protobuf gRPC for device RPC; CRDTs via Yjs and Rustler NIFs; gen_server per-document collaboration; Phoenix.PubSub with libcluster; hot upgrades using relup/appup; DeployEx release orchestration; Livebook connecting GRiSP boards; Gleam for microcontrollers; Unicode handling in SQL virtualization
| title | author | length |
|---|---|---|
| LiveView and Web Components | Chris Nelson | 24m |
Argues that Web Components extend LiveView by becoming first‑class UI elements and shows a working autocomplete + Google Maps demo. LiveElements converts… | ||
| Nerves ❤️ Flutter | Digit | 29m |
SmartRent ports Flutter/Dart UIs onto Nerves Linux hubs, shipping a touchscreen thermostat/hub (Fusion3) that runs locally and controls Z-Wave devices; the… | ||
| CRDTs and the BEAM | Kevin Barrett, Sloane Perrault | 28m |
screen.garden implements Markdown collaboration by embedding Yjs (Y_RS/wires) wrapped via a Rustler NIF on clients and mirroring docs as processes on the BEAM… | ||
| Learn You Some Agroecology for Great Good! | John Mandalakas | 25m |
Survey of agricultural history claims figs as possibly the first domesticate and credits the Haber process for synthetic nitrogen (mostly using natural gas… | ||
| Your Official Elixir Expert | Mitchell Hanberg | 43m |
Three independent Elixir LSPs (Elixir LS, Next LS, Lexical) are merging into a single official implementation called Expert that combines multi-node… | ||
| Test Automation – Get the BEAM into an Organization | Magnus Sedlacek | 26m |
Argues that test automation is the practical on‑ramp for BEAM adoption by showing a telecom case where common_test and Erlang replace a fractured C and Ruby… | ||
| Cracking the PubSub Code. Elixir's Key to Scalable Real-Time Systems | Alvaro Callero | 23m |
Shows a concrete PubSub setup using Phoenix.PubSub with a GenServer state store that broadcasts an updated car count across two nodes connected by libcluster… | ||
| Working with Multilingual Data in SQL Databases through Erlang | Gourab Mitra | 27m |
A SQL virtualization engine written in Erlang translates Teradata and Oracle queries to BigQuery Azure or Postgres and proves that correct Unicode handling… | ||
| Detective Hat: Investigating Production Issues | Adrian Dunston, Jenny Bramble | 41m |
A pragmatic checklist for chasing production bugs: five core laws—all production issues are failures of process, believe the problem until you prove it isn’t… | ||
| Livebook Meets GRiSP | Peer Stritzinger | 42m |
Livebook runs on a cloud node and connects to GRiSP boards to run code remotely, demoing LED control and sensors on two boards (one in Costa Rica) and a cloud… | ||
| Acceptable Upgrades | Eric Saxby | 28m |
Legacy DB forces tests to use a full production dump so initial tests require at least 18 records and run painfully slow. The remedy is concrete: draw the… | ||
| Bring Back the Power of BEAM Deployments with DeployEx | Thiago Esteves | 36m |
A tool called DeployEx is an Elixir-based orchestrator that runs on the host, watches an S3 bucket for new mix release packages, and performs either full… | ||
| Crafting Fully Custom Code Generators | Aidan Obley, Jeffrey Matthias | 26m |
Custom Elixir generators take an Elixir input file (keyword lists or structs), run a preprocessing digestor that pulls from legacy source code, an Oracle… | ||
| Live voices with LiveView | Lars Wikman | 39m |
A working local voice assistant pipeline is built that runs a small LLM (135M params) plus Whisper for transcription, a lightweight VAD, and a Kokoro-like TTS… | ||
| Battling Memory Leaks: Tales from the Trenches at WhatsApp | Miriam Pena | 44m |
WhatsApp describes concrete fixes and diagnostics: they run per-second instrumentation written to file, a memory outlier detector that dumps state or forces… | ||
| Open Riak how we gained from leaning on the BEAM | Martin Sumner | 40m |
Migrated 20k lines of C++ to Erlang and redesigns give measurable wins: a 150M-object cluster on eight modest EC2 nodes (~$4k/month) stays stable at 100% CPU… | ||
| Choosing an effective testing structure | Jason Axelson | 22m |
Advocates writing tests that fail with helpful messages, puts expected values on the right, and uses Machete to match only relevant fields; prefers async tests… | ||
| Microcontrollers with Gleam | Ray De Los Santos | 25m |
Framework for running Gleam on tiny devices using AtomVM with working prototypes on ESP32 and Raspberry Pico, showing actor-based concurrency, GPIO and I2C… | ||
| Keynote: The Socio-technical Elements that Make Good Platforms | Charity Majors, Fred Hebert | 01h 02m |
Platform engineering works when teams own code in production, are on call for it, and get tooling and permission to change things; Honeycomb moves from one… | ||
| Keynote: AI for worker collective action | Saiph Savage | 41m |
A web plugin tracker logs paid and unpaid time and shows many platform workers fall below minimum wage; LLM-assisted collective analysis clusters worker data… | ||
| OTP Update | Ingela Anderton Andin | 19m |
OTP 27 adds Process Labels for richer crash reports and Observer listings, switches docs from XML to Markdown with triple-quoted API examples, and accepts a… | ||
| Keynote: Fault Tolerant Machine Learning Operations | Chelsea Troy | 51m |
Defines three maturity levels (0–2) and shows moving from ad hoc local models to automated retraining, architecture search and orchestration using DAGs and… | ||
| Elixir Update | Eric Meadows–Jönsson | 26m |
Recent stable work ships as 1.18 release and targets a mid‑year 1.19 with the bulk of effort on the type system: a set‑theoretic, gradual, sound design with a… | ||
| Keynote: Designing LLM Native systems | Sean Moriarity | 42m |
Argues that chatbots often harm products and recommends avoid building chat interfaces in most cases, instead using structured outputs (Instructor library… | ||
| Smarter Apps with Ash and GenAI | Josh Price | 39m |
Ash resources become a machine-readable API that an LLM can call: a ~500-line Ash mCP server exposes actions as tools (reads, search, create) via OpenAI… | ||
| Fullstack Gleam: Static Types on the BEAM, and Javascript You'll Love | Brett Kolodny | 28m |
Gleam is a statically typed language that compiles to Erlang and JavaScript and here it builds a full-stack blog: front-end UI with luster (Elm-style… | ||
| nothing matches | ||