Senior Full-Stack AI Engineer

Job Title: Senior Full-Stack AI Engineer (Backend-leaning)

Location: Remote · Type: Full-time contract · Duration: 3 months (estimated), tied to project completion · Level: 3+ years building and running production software

About the role

Entropik Labs is hiring a senior engineer to lead the backend of a production mobile product with a real-time, AI-powered backend. The system streams audio and text to and from mobile clients under a tight latency budget. Heavier processing runs on background workers, and the product runs on a subscription model.

The backend is built on FastAPI, alongside WebSockets, PostgreSQL, Redis, a task queue and AWS. It integrates third-party APIs for LLMs, speech, translation, auth, storage and payments. The mobile client is built in Flutter.

This is a full-time contract role tied to the completion of the project, estimated at 3 months. You will lead the engineering team through delivery, so you'll be setting direction as well as writing code.

What you'll own

• Technical leadership. You'll lead the engineering team for the project: set standards and architecture, plan and prioritise work, review code, mentor junior engineers, and be accountable for delivery.

• Real-time systems. You'll own a bidirectional WebSocket service that streams audio and text between mobile clients and a realtime model API. It has to handle reconnects, superseded connections, heartbeats, partial messages and provider outages.

• Async processing. You'll build queue-backed workers with idempotency, deduplication, retries and redelivery designed in from the start.

• Payments and subscriptions. This covers App Store and Google Play in-app purchases, server-side receipt validation, webhooks and entitlements. You'll handle trials, renewals, refunds and grace periods.

• Data modelling and migrations. You'll own a relational schema that answers core queries cheaply, evolves without downtime, and runs on pooled, managed Postgres.

• Caching and rate limiting. You'll use Redis as cache, counter, lock and broker, and decide what data is authoritative and what is derived.

• Third-party integrations. You'll wrap external providers so that switching one is a config change and their failures degrade the product gracefully.

• Production on AWS. You'll own containers, deploys, networking, IAM, managed databases, queues, storage, structured logging, tracing and alerting. You'll debug performance and cost problems from evidence.

• API design. You'll manage multiple API versions side by side and plan how clients migrate between them.

• Handover. You'll leave the system well documented, with runbooks and a clean handover at the end of the project.

What we need from you

• Leadership. You have proven experience leading engineering teams or projects: owning architecture decisions, breaking work into milestones, reviewing and mentoring, and keeping delivery on track. You flag risks early and communicate clearly with both engineers and stakeholders.

• Production experience. You have 5+ years of shipping and operating backend services in production, including on-call, incident response, and debugging issues that only appear in production.

• Python and FastAPI. You have built and run production services in FastAPI, including async endpoints, Pydantic models, dependency injection, middleware, background tasks and WebSocket endpoints. You're fluent with asyncio concurrency, cancellation, timeouts and backpressure.

• AWS expertise. You have hands-on production experience with ECS, EKS, EC2 or Lambda, plus RDS/Aurora, ElastiCache, SQS/SNS, S3, IAM, VPC networking and CloudWatch. You design for security, reliability and cost.

• Database depth. You have strong PostgreSQL skills: schema design, indexing, query plans, transactions, connection pooling and zero-downtime migrations. You also know backups, point-in-time recovery and read replicas.

• Payments. You have built or maintained payment or subscription systems in production, such as App Store or Google Play billing, or Stripe. You understand webhooks, idempotency keys, entitlement state, reconciliation, and events that arrive duplicated or out of order.

• Real-time or streaming experience. This could be WebSockets, SSE, gRPC streaming, audio/video pipelines, chat, or collaborative editing.

• Queues and background jobs. You know at-least-once delivery, idempotent handlers, poison messages and scheduled jobs.

• Full-stack range. You've built or maintained a web or mobile client, and you can work closely with mobile developers on API design.

• Deployment ownership. You handle Docker, CI/CD, environment and secrets management, and safe rollouts and rollbacks.

• Delivery under a deadline. You've shipped against fixed scopes and timelines, and you can get productive in an existing codebase quickly.

Nice to have

• Infrastructure as code (Terraform, AWS CDK or CloudFormation).

• Production work with LLM APIs: streaming, structured output, prompt and model versioning, evaluation, and cost and latency control.

• Speech processing: ASR, TTS, audio encoding, PCM/MP3 handling, voice activity detection.

• NLP fundamentals, or experience with libraries like spaCy.

• Experience with Supabase or Firebase, especially their auth and storage models.

• Mobile-adjacent backend work: offline behaviour, flaky networks, payload size.

• Load testing, performance profiling, or owning an SLO.

• Experience on a consumer subscription product.

What we look for

We want to see that you've taken a system from a working prototype to something people rely on, and led others while doing it. Be ready to walk us through the specific decisions that got it there, including one you would take back.