Backend Engineer (Elixir/OTP) - Full-time - Worldwide

About the Role

We're building an innovative agentic code remediation platform that uses AI to fix code quality issues in real-time. Our system integrates with GitHub Actions, SonarQube, and LLMs to automatically create pull requests that fix code issues, creating a self-healing CI/CD pipeline.

We're looking for a Backend Engineer with deep Elixir/OTP expertise to build the orchestration layer of our distributed system. You'll focus on creating a fault-tolerant, scalable platform using Phoenix, Absinthe GraphQL, and BEAM's unique capabilities.

The Platform Architecture

Our automated remediation agent consists of:
● Elixir/Phoenix orchestrator (your focus) - Event processing, safety enforcement,
and API layer
● Python microservices - Git operations and LLM integration
● Fire-and-forget architecture - Async processing with natural Git-driven iterations
● BEAMops deployment - Leveraging BEAM's operational characteristics alongside Kubernetes
Your Domain:
● Phoenix web framework with LiveView dashboards
● Absinthe GraphQL API with real-time subscriptions
● GenStage/Broadway for event processing pipelines
● Supervised OTP applications with fault tolerance
● Distributed BEAM nodes for horizontal scaling

What You'll Do

Core Responsibilities
● Design and implement OTP applications with proper supervision trees for fault tolerance
● Build Phoenix APIs and Absinthe GraphQL schemas with real-time subscriptions
● Develop GenStage pipelines for processing remediation events with backpressure
● Create GenServer-based state machines for managing remediation workflows
● Implement distributed rate limiting and circuit breakers using OTP patterns
● Design event-driven architectures leveraging Phoenix PubSub and EventBus
● Build Phoenix LiveView components for real-time monitoring dashboards
● Optimize for BEAM's strengths - massive concurrency and fault tolerance

Example Projects You'll Own

1. Supervision Architecture: Design supervisor hierarchies that gracefully handle failures in external services
2. GraphQL Subscriptions: Implement real-time updates for PR remediation status using Absinthe subscriptions
3. Event Processing Pipeline: Build GenStage/Broadway pipelines handling thousands of concurrent remediations
4. Distributed State Management: Implement distributed state using OTP patterns, ETS, and potentially Mnesia
5. BEAMops Infrastructure: Set up hot code reloading, distributed tracing, and BEAM-specific monitoring
6. Rate Limiting System: Build a distributed rate limiter using GenServer and ETS
7. Config Hot Reloading: Implement configuration management with runtime updates

Technical Requirements

● 3+ years production Elixir development experience
● Deep understanding of OTP principles: supervisors, GenServers, and fault tolerance
● Production experience with Phoenix Framework and Phoenix LiveView
● Experience building GraphQL APIs with Absinthe
● Proficiency in designing supervision trees and failure recovery strategies
● Understanding of BEAM VM characteristics and operational concerns
● Experience with GenStage, Flow, or Broadway for data processing
● Testing expertise with ExUnit, property-based testing, and distributed system testing
● BEAMops experience: Hot upgrades, distributed deployments, BEAM clustering
● Distributed Elixir: Experience with distributed Erlang/Elixir applications
● Phoenix LiveView advanced patterns and JavaScript hooks
● Ecto with PostgreSQL for metadata storage
● Telemetry and OpenTelemetry for observability
● Experience with libcluster for automatic node discovery
● NIFs/Ports for integrating with external services

Technical Stack You'll Work With

Core Elixir Stack:
● Elixir 1.14+ / Erlang/OTP 25+
● Phoenix 1.7+ with LiveView
● Absinthe GraphQL
● GenStage/Broadway
● Ecto with PostgreSQL
● Oban for background jobs
Infrastructure:
● BEAMops practices (hot code upgrades, distributed deployments)
● Kubernetes with BEAM-aware operators
● Docker for containerization
● GitHub Actions for CI/CD
● Prometheus + Grafana for metrics

What Makes You a Great Fit

You're an OTP Expert
● Design systems thinking "let it crash" philosophy
● Know when to use GenServer vs Agent vs Task
● Understand supervision strategies and when to apply each
● Can reason about distributed system guarantees in BEAM
You Embrace Functional Programming
● Think in terms of data transformations and pipelines
● Write composable, testable functions
● Understand immutability benefits for concurrent systems
You Build for Resilience
● Design for failure as the normal case
● Implement circuit breakers and bulkheads naturally
● Use supervision trees to isolate failures
● Know how to handle network partitions
You Understand BEAM's Superpowers
● Leverage massive concurrency (millions of processes)
● Use message passing for decoupling
● Implement soft real-time guarantees
● Know when BEAM is the right tool vs when it's not