Production ML Engineer for Mobile Trading App (Hybrid)

Endava Limited; Greater London; Hybrid; GBP 75,000 - 120,000; Full time

Benefits offered by this job

Hybrid work hours

Career development programs

Wellbeing programme

Referral bonus

Wellbeing apps

Job summary

Endava is seeking ML Engineers to productionalize machine learning models for our new mobile trading app. You will deploy, operate, and monitor models, and build data pipelines the models rely on.

You will deploy models as low-latency services, orchestrate data flows with Airflow, and integrate external feeds into a scalable data warehouse. This role emphasizes reliability, observability, and handover-ready documentation.

Qualifications

• 4+ years in software or ML engineering with models running in production.
• Strong Python; model serving via REST/gRPC APIs, Docker and Kubernetes or equivalent.
• AWS experience (e.g. S3, compute, managed ML services).
• Data pipelines and orchestration (Airflow or similar), including third-party API integration.
• CI/CD, infrastructure-as-code and observability tooling.
• Experience in high-availability environments where outages affect customers.

Responsibilities

• Deploy models as reliable, low-latency services consumed by the mobile app backend.
• Build and schedule feature pipelines, including daily refresh of customer-level scores.
• Integrate an external market-news feed and product analytics data into our data warehouse.
• Implement model monitoring, drift detection, alerting and retraining workflows; establish deployment patterns, CI/CD and testing standards for ML in production.
• Document everything needed for long-term ownership and deliver a handover plan.

Skills

Python

REST APIs

CI/CD

Data Pipelines

Model monitoring

High availability

AI/ML concepts

Cloud experience

Tools

Docker

Kubernetes

Airflow

AWS

Job description

Endava is seeking ML Engineers to productionalize machine learning models for our new mobile trading app. You will deploy, operate, and monitor models, and build data pipelines the models rely on.

You will deploy models as low-latency services, orchestrate data flows with Airflow, and integrate external feeds into a scalable data warehouse. This role emphasizes reliability, observability, and handover-ready documentation.