Senior AI Platform Engineer – Full-Stack & LLM Systems

Protagonist; Washington (District of Columbia); On-site; USD 95,000 - 141,000; Full time

Job summary

Protagonist is seeking a Senior Full Stack Engineer (AI Platform) to own production systems end-to-end—from backend data infrastructure to analyst-facing UIs—and to apply modern AI capabilities within a robust software platform.

The ideal candidate excels as a full-stack engineer, has experience building AI- and LLM-powered applications in production, and can navigate ambiguity while delivering scalable, reliable software.

Qualifications

• U.S. person eligible to work on government contracts; SECRET clearance preferred.
• 5+ years building production software with ownership of full-stack systems.
• Expert-level Python with Django and PostgreSQL experience.
• Strong TypeScript/React skills for data-heavy interfaces.
• Experience building LLM applications: agents, tool calling, RAG, evaluation.

Responsibilities

• Own features end-to-end across the platform: Django REST APIs, React UI, pipelines, and LLM services.
• Collaborate on architecture and roadmap; turn product vision into system designs.
• Build LLM-powered systems on AWS Bedrock; implement tool usage and retrieval flows.
• Develop data and ML workflows using Metaflow on Argo/Kubernetes.
• Design PostgreSQL schemas, OpenSearch indexing, and API contracts.

Skills

Python

Django

PostgreSQL

TypeScript

React

LLM apps

OpenSearch

Kafka

Celery/Redis

AWS

Docker

Kubernetes

Git

CI/CD

Terraform

Education

Bachelor's degree in Computer Science or Computer Engineering

Master's degree in Computer Science or Computer Engineering

Tools

OpenSearch

Elasticsearch

Terraform

Job description

Protagonist is seeking a Senior Full Stack Engineer (AI Platform) to own production systems end-to-end—from backend data infrastructure to analyst-facing UIs—and to apply modern AI capabilities within a robust software platform.

The ideal candidate excels as a full-stack engineer, has experience building AI- and LLM-powered applications in production, and can navigate ambiguity while delivering scalable, reliable software.