Elixir Wizards

Elixir creator José Valim returns to the podcast to unpack the latest developments in Elixir’s set-theoretic type system and how it is slotting into existing code without requiring annotations. We discuss familiar compiler warnings, new warnings based on inferred types, a phased rollout in v1.19/v1.20 that preserves backward compatibility, performance profiling the type checks across large codebases, and precise typing for maps as both records and dictionaries. José also touches on CNRS academic collaborations, upcoming LSP/tooling enhancements, and future possibilities like optional annotations and guard-clause typing, all while keeping Elixir’s dynamic, developer-friendly experience front and center. Key topics discussed in this episode: Set-theoretic typing (union, intersection, difference) Compiler-driven inference with zero annotations Phased rollout strategy in 1.19 and 1.20 Performance profiling for large codebases Map typing as records and dictionaries Exhaustivity checks and behavioral typing in GenServers Language Server Protocol & tooling updates Future optional annotations and guard-clause typing CNRS collaboration for theoretical foundations Clear error messages and false-positive reduction Community-driven feedback and iterative improvements Links mentioned: https://github.com/elixir-nx https://livebook.dev/ https://hexdocs.pm/phoenix_live_view/Phoenix.LiveView.html https://hexdocs.pm/elixir/main/gradual-set-theoretic-types.html https://hexdocs.pm/dialyxir/0.4.0/readme.html https://remote.com/ Draw the Owl meme: https://i.imgur.com/rCr9A.png https://dashbit.co/blog/data-evolution-with-set-theoretic-types https://hexdocs.pm/ecto/Ecto.html https://github.com/elixir-lsp/elixir-lsSpecial Guest: José Valim.

Zack Kayser, Staff Software Engineer at cars.com, joins Elixir Wizards Sundi Myint and Charles Suggs to discuss how Cars.com adopted a server-driven UI (SDUI) architecture powered by Elixir and GraphQL to deliver consistent, updatable interfaces across web, iOS, and Android. We explore why SDUI matters for feature velocity, how a mature design system and schema planning make it feasible, and what it takes, culturally and technically, to move UI logic from client code into a unified backend. Key topics discussed in this episode: SDUI fundamentals and how it differs from traditional server-side rendering GraphQL as the single source of truth for UI components and layouts Defining abstract UI components on the server to eliminate duplicate logic Leveraging a robust design system as the foundation for SDUI success API-first development and cross-team coordination for schema changes Mock data strategies for early UI feedback without breaking clients Handling breaking changes and hot-fix deployments via server-side updates Enabling flexible layouts and A/B testing through server-controlled ordering Balancing server-driven vs. client-managed UI Iterative SDUI rollout versus “big-bang” migrations in large codebases Using type specs and Dialyxir for clear cross-team communication Integration testing at the GraphQL layer to catch UI regressions early Quality engineering’s role in validating server-driven interfaces Production rollback strategies across web and native platforms Considerations for greenfield projects adopting SDUI from day one Zack and Ethan's upcoming Instrumenting Elixir Apps book Links mentioned: https://cars.com https://github.com/absinthe-graphql/absinthe Telemetry & Observability for Elixir Apps Ep: https://youtu.be/1V2xEPqqCso https://www.phoenixframework.org/blog/phoenix-liveview-1.0-released https://hexdocs.pm/phoenix_live_view/assigns-eex.html https://graphql.org/ https://tailwindcss.com/ https://github.com/jeremyjh/dialyxir https://github.com/rrrene/credo GraphQL Schema https://graphql.org/learn/schema/ SwiftUI https://developer.apple.com/documentation/swiftui/  Kotlin https://kotlinlang.org/ https://medium.com/airbnb-engineering/a-deep-dive-into-airbnbs-server-driven-ui-system-842244c5f5 Zack’s Twitter: https://x.com/kayserzl/ Zack’s LinkedIn: https://www.linkedin.com/in/zack-kayser-93b96b88 Special Guest: Zack Kayser.

Sonny Scroggin, a member of the Rustler Core Team and an expert in high-performance applications, joins Sundi Myint and Charles Suggs for an engaging discussion. They explore the seamless integration of Rust and Elixir, focusing on Native Implemented Functions (NIFs) for CPU-intensive tasks. Sonny shares insights on when to use NIFs versus other approaches, emphasizing Rust's memory safety. Exciting future developments for Rustler, including asynchronous NIF interfaces, are highlighted, alongside the enduring strengths of Elixir's dynamic model.

Sean Moriarity, co-creator of the Nx project and author of Machine Learning in Elixir, shares his journey from military to tech, focusing on large language models. He discusses the evolving Elixir ML landscape, the strategic choice between off-the-shelf models and custom solutions, and how to effectively integrate Python libraries. Sean highlights Elixir's strengths in distributed computing for ML tasks and offers practical advice on starting ML projects, emphasizing the use of motivating use cases and domain-specific languages to enhance AI workflows.

Mark Ericksen, creator of the Elixir LangChain framework, joins the Elixir Wizards to talk about LLM integration in Elixir apps. He explains how LangChain abstracts away the quirks of different AI providers (OpenAI, Anthropic’s Claude, Google’s Gemini) so you can work with any LLM in one more consistent API. We dig into core features like conversation chaining, tool execution, automatic retries, and production-grade fallback strategies. Mark shares his experiences maintaining LangChain in a fast-moving AI world: how it shields developers from API drift, manages token budgets, and handles rate limits and outages. He also reveals testing tactics for non-deterministic AI outputs, configuration tips for custom authentication, and the highlights of the new v0.4 release, including “content parts” support for thinking-style models. Key topics discussed in this episode: • Abstracting LLM APIs behind a unified Elixir interface • Building and managing conversation chains across multiple models • Exposing application functionality to LLMs through tool integrations • Automatic retries and fallback chains for production resilience • Supporting a variety of LLM providers • Tracking and optimizing token usage for cost control • Configuring API keys, authentication, and provider-specific settings • Handling rate limits and service outages with degradation • Processing multimodal inputs (text, images) in Langchain workflows • Extracting structured data from unstructured LLM responses • Leveraging “content parts” in v0.4 for advanced thinking-model support • Debugging LLM interactions using verbose logging and telemetry • Kickstarting experiments in LiveBook notebooks and demos • Comparing Elixir LangChain to the original Python implementation • Crafting human-in-the-loop workflows for interactive AI features • Integrating Langchain with the Ash framework for chat-driven interfaces • Contributing to open-source LLM adapters and staying ahead of API changes • Building fallback chains (e.g., OpenAI → Azure) for seamless continuity • Embedding business logic decisions directly into AI-powered tools • Summarization techniques for token efficiency in ongoing conversations • Batch processing tactics to leverage lower-cost API rate tiers • Real-world lessons on maintaining uptime amid LLM service disruptions Links mentioned: https://rubyonrails.org/ https://fly.io/ https://zionnationalpark.com/ https://podcast.thinkingelixir.com/ https://github.com/brainlid/langchain https://openai.com/ https://claude.ai/ https://gemini.google.com/ https://www.anthropic.com/ Vertex AI Studio https://cloud.google.com/generative-ai-studio https://www.perplexity.ai/ https://azure.microsoft.com/ https://hexdocs.pm/ecto/Ecto.html https://oban.pro/ Chris McCord’s ElixirConf EU 2025 Talk https://www.youtube.com/watch?v=ojL_VHc4gLk Getting started: https://hexdocs.pm/langchain/getting_started.html https://ash-hq.org/ https://hex.pm/packages/langchain https://hexdocs.pm/igniter/readme.html https://www.youtube.com/watch?v=WM9iQlQSF_g @brainlid on Twitter and BlueSkySpecial Guest: Mark Ericksen.

Connor Rigby joins the Elixir Wizards to talk about Blue Heron BLE (Bluetooth Low Energy) support for Elixir apps. Blue Heron implements the BLE specs in pure Elixir, leveraging binary pattern matching and concurrent message processing to handle Bluetooth protocols. Unlike most solutions that require C ports or NIFs, Blue Heron runs entirely in user space, so it works seamlessly in both Nerves-based embedded projects and (eventually) desktop Elixir applications. We discuss how Nerves development differs from building Phoenix apps. Connor shares challenges he's experienced with hardware compatibility, where some chips only partially implement the spec, and he discusses the surprisingly deep (but sometimes incomplete) world of BLE device profiles. His tip for anyone entering the BLE space: read the official spec instead of trusting secondhand blog posts. Tools like Nerves LiveBook give you hands-on examples, so you can get a BLE prototype running on a Raspberry Pi and your phone in no time. Key topics discussed in this episode: Blue Heron origins and “bird” naming convention BLE vs. Bluetooth Classic: core differences Pure Elixir implementation—no C dependencies Binary pattern matching for packet parsing Hardware transport options: UART, SPI, USB, SDIO GenServer patterns in Nerves vs. Phoenix Linux requirement and power-consumption trade-offs GATT (Generic Attribute Table) implementation patterns SQLite integration for Nerves apps Hardware chip quirks and spec compliance Manufacturer-specific commands and workarounds BLE device profiles and spec gaps Security Management Profile (SMP) for encryption Device connection and pairing workflows Web vs. embedded development differences Where to get started: hardware recommendations and docs Links mentioned: https://github.com/ConnorRigby/ https://github.com/blue-heron/ https://nerves-project.org/ BLE (Bluetooth Low Energy) https://en.wikipedia.org/wiki/Bluetooth_Low_Energy https://developer.apple.com/ibeacon/ https://learnyousomeerlang.com/building-otp-applications Linux https://www.linux.org/ HCI (Host Controller Interface) https://en.wikipedia.org/wiki/Host_controller_interface Circuits UART Library https://hexdocs.pm/circuits_uart/readme.html SPI (Serial Peripheral Interface) https://github.com/elixir-circuits/circuits_spi SDIO (Secure Digital Input Output https://en.wikipedia.org/wiki/SDIO Raspberry Pi https://www.raspberrypi.com/ Coral SoM Dev Board https://coral.ai/products/dev-board/ BeagleBone Single-Board Linux Computer https://www.beagleboard.org/boards/beaglebone-black https://www.bluetooth.com/bluetooth-resources/intro-to-bluetooth-gap-gatt/ Genservers https://hexdocs.pm/elixir/1.12/GenServer.html https://hexdocs.pm/ecto/Ecto.html https://github.com/elixir-sqlite/ecto_sqlite3 https://github.com/nerves-livebook/nerves_livebookSpecial Guest: Connor Rigby.

Isaac Yonemoto, creator of the Zigler library and CEO of Videla Labs, discusses his venture into combining Zig's performance with Elixir's capabilities using Native Implemented Functions (NIFs). He explains Zigler's auto-generated Elixir-to-Zig bridge and its three execution modes: normal, dirty, and threaded. Isaac shares insights on his biotech startup that leverages Elixir for bioinformatics and highlights the benefits of Zig in hardware acceleration for large language models. He also touches on the importance of fault tolerance in low-level coding practices.

Marc Lainez, Thibault Poncelet, and Loïc Vigneron, software engineers at Spin42, delve into their innovative project retrofitting a 2007 VW Polo with an Open Vehicle Control System using Elixir and Nerves. They share their journey of integrating modern tech into classic cars, emphasizing environmental sustainability and community engagement. The team discusses overcoming challenges like reverse engineering CAN messages, creating a vehicle management system, and the fun of collaborating on an open-source platform. Potential applications for their system range from boats to construction vehicles.

In this engaging discussion, Jim Freeze, organizer of ElixirConf and creator of the Horizon library, shares his journey from Ruby to Elixir. He discusses the evolution of ElixirConf, shaped by community input and the challenges of COVID-19. Jim details Horizon's design for deploying Elixir apps on FreeBSD, emphasizing simplicity and minimal dependencies. He highlights FreeBSD's security and performance benefits, encouraging developers to take charge of their deployment stacks. Community engagement and open-source contributions are underlined as essential for growth and innovation.

Zack Kayser and Ethan Gunderson, Software Engineers at Cars Commerce, join the Elixir Wizards to share their expertise on telemetry and observability in large-scale systems. Drawing from their experience at Cars.com—a platform handling high traffic and concurrent users—they discuss the technical and organizational challenges of scaling applications, managing microservices, and implementing effective observability practices. The conversation highlights the pivotal role observability plays in diagnosing incidents, anticipating system behavior, and asking unplanned questions of a system. Zack and Ethan explore tracing, spans, and the unique challenges introduced by LiveView deployments and WebSocket connections. They also discuss the benefits of OpenTelemetry as a vendor-agnostic instrumentation tool, the significance of Elixir’s telemetry library, and practical steps for developers starting their observability journey. Additionally, Zack and Ethan introduce their upcoming book, Instrumenting Elixir Applications, which will offer guidance on integrating telemetry and tracing into Elixir projects. Topics Discussed: Cars.com’s transition to Elixir and scaling solutions The role of observability in large-scale systems Uncovering insights by asking unplanned system questions Managing high-traffic and concurrent users with Elixir Diagnosing incidents and preventing recurrence using telemetry Balancing data collection with storage constraints Sampling strategies for large data volumes Tracing and spans in observability LiveView’s influence on deployments and WebSocket behavior Mitigating downstream effects of socket reconnections Contextual debugging for system behavior insights Observability strategies for small vs. large-scale apps OpenTelemetry for vendor-agnostic instrumentation Leveraging OpenTelemetry contrib libraries for easy setup Elixir’s telemetry library as an ecosystem cornerstone Tracing as the first step in observability Differentiating observability from business analytics Profiling with OpenTelemetry Erlang project tools The value of profiling for performance insights Making observability tools accessible and impactful for developers Links Mentioned https://www.carscommerce.inc/ https://www.cars.com/ https://hexdocs.pm/telemetry/readme.html https://kubernetes.io/ https://github.com/ninenines/cowboy https://hexdocs.pm/bandit/Bandit.html https://hexdocs.pm/broadway/Broadway.html https://hexdocs.pm/oban/Oban.html https://www.dynatrace.com/ https://www.jaegertracing.io/ https://newrelic.com/ https://www.datadoghq.com/ https://www.honeycomb.io/ https://fly.io/phoenix-files/how-phoenix-liveview-form-auto-recovery-works/ https://www.elastic.co/ https://opentelemetry.io/ https://opentelemetry.io/docs/languages/erlang/ https://opentelemetry.io/docs/concepts/signals/traces/ https://opentelemetry.io/docs/specs/otel/logs/ https://github.com/runfinch/finch https://hexdocs.pm/telemetry_metrics/Telemetry.Metrics.html https://opentelemetry.io/blog/2024/state-profiling https://www.instrumentingelixir.com/ https://prometheus.io/ https://www.datadoghq.com/dg/monitor/ts/statsd/ https://x.com/kayserzl https://github.com/zkayser https://bsky.app/profile/ethangunderson.com  https://github.com/open-telemetry/opentelemetry-collector-contribSpecial Guests: Ethan Gunderson and Zack Kayser.