Backend ML Engineer at Robyn AI

Robyn is not just an AI app — she’s your emotionally intelligent companion. A trusted mirror. A guide. A new kind of OS for your emotional life. We're building the world’s first emotional intelligence layer for AI. You’ll be building the backend infrastructure that powers all of it:

• Conversations, memory, and real-time personalization
• Voice + chat interface
• Scalable infra for emotional intelligence
• Secure and fast APIs for our iOS app
• A robust ML inference and fine-tuning pipeline

You’ll be the technical backbone of Robyn — designing and shipping fast, scalable, emotionally aware systems while collaborating closely with iOS, product, and AI teams. This is a rare opportunity to define the foundations of emotionally intelligent AI. Everything beyond the core LLM — memory, emotional layer, and relational engine — is built fully in-house. The backend engineer will help architect the systems that make Robyn feel human: writing the foundational codebase for the next wave of AI — one that feels, remembers, and connects. What You'll Do

• Backend & Infra Ownership
• You'll work and add to our C# / .NET / ASP.NET backend api layer (have experience in this or something similar like Java/Spring and can learn quickly) and progressively add many Python microservices (FastAPI or AWS Lambda) with modular, AI-native architecture in mind to build our intelligence layer. Deploy models and setup some ml training pipelines.
• Understand dependency injection, Strategy Pattern, inversion of control, and other best practices for code maintainability
• Own the full backend surface area — auth, APIs, infra, orchestration — and design all of your features for scale and velocity.
• Build and maintain REST and GraphQL APIs consumed by our iOS client; low-latency, resilient, and well-instrumented.
• Architect a microservice-style ML model serving backend deployed via Docker containers or AWS Lambda (SnapStart), backed by async eventing and pub/sub where needed.
• Own CI/CD, rollback strategies, logging, error handling — the backend is your domain, end-to-end.
• AI & ML Systems Integration
• Architect and manage existing vector DB (PgVector) and potentially add more to power retrieval-augmented generation, evolving memory, and personalization.
• Build tools and add to our custom memory pipelines tied to user context, embeddings, and interaction history.
• Integrate and scale inference with OpenAI, Claude, Llama, or other models. Build wrappers, manage caching, fallbacks, and prompt routing logic.
• Own emotion and sentiment tagging workflows; plug in APIs or run lightweight classifiers in-line.
• Maintain API orchestration layer with 3rd party model providers (OpenAI, ElevenLabs, etc.).
• Cloud Infra, DevOps, and Data Stack
• Manage our AWS infrastructure and add to our current stack with new innovate technologies: Lambda, ECS, S3, RDS (Postgres), CloudFront, IAM, Route53 — you’ll be the one making the call on architecture and trade-offs.
• Be able to use search databases like OpenSearch
• Infra-as-code with Terraform. Pipelines through GitHub Actions.
• Full observability: metrics, structured logging, tracing, alerting — Open Telemetry, Sentry, Grafana, Cloudwatch, etc.
• Optimize latency across API surface, tune Postgres indexes, add Redis caching layers to many of your services and pub/sub or streaming for near-instantaneous data sync.
• Set up and secure infra for SOC-2 readiness: access controls, data lifecycle policies, encrypted storage.
• Personalization & Emotional Intelligence Layer
• Design and ship emotion-aware backend systems that update in real time based on user behavior.
• Implement custom memory engines — user embeddings, experience graphs, emotional scores — and build APIs that adapt over time.
• Work with product and AI to tune behavior of Robyn based on user feedback, emotion logs, memory history, and interaction loops.
• All personalization logic

You’re Probably Right for This If:

• You have 6 - 15 years of experience in backend or full-stack development; building 0-1 products or teams in a startup environment
• You’ve shipped entire production backends at high-growth early-stage startups — you know how to move fast and still write code you don’t hate six months later.
• You’ve integrated or scaled LLM-based products — bonus if you’ve done it with emotion, memory, or personalization layers.
• You care about systems thinking, fast response times, clean abstractions, and building infra that won’t fall over under load.
• You’ve done the zero-to-one and can hold both the product in your head and the infra in your hands.
• You’re able to figure things out quickly and dive in wherever needed. There’s no “that’s not my job” here.
• You're allergic to bloated teams — you’d rather build it right yourself than manage a swarm of mid devs doing it wrong.

Compensation Range: $150K - $250K