Skip to content
mimi

Senior AI Systems Architect

DarkRoast Design

Peace River · On-site Contract Senior 3w ago

About the role

About

We’re building an AI intelligence infrastructure embedded inside a B2B platform used by consumer brands to manage creative work, assets, and workflows.

This infrastructure is designed to support a wide range of intelligent and intuitive product features, including (but not limited to) conversational interfaces, image generation, contextual recommendations, automation, analysis, and decision support.

This is not a chatbot project.

This is not a ChatGPT wrapper.

Role

We’re looking for a senior AI Systems Architect to design the foundational intelligence layer that powers these capabilities in a way that is reliable, evolvable, and cost‑controlled.

Responsibilities

You will own the design of the AI intelligence layer and this includes:

  • Designing a clear intelligence architecture that supports multiple AI‑driven features
  • Defining a robust memory taxonomy (short‑term, long‑term, client‑level, org‑level, and meta memory)
  • Designing RAG‑based retrieval systems that serve different product use cases
  • Defining retrieval + reasoning flows that adapt to intent, context, and confidence
  • Establishing failure handling and guardrail strategies for production AI systems
  • Designing idempotent AI workflows for actions, updates, and automation
  • Establishing cost controls, usage boundaries, and observability
  • Designing evaluation and feedback loops so intelligence improves through use

Deliverables

  • Memory schemas and retrieval strategies
  • Interface contracts and reference patterns for implementation
  • A Phase‑2 build plan outlining how intelligence evolves post‑MVP

What You'll Not Do

  • Building UI or frontend features
  • Acting as a generalist full‑stack developer
  • Shipping one‑off demos or prototypes
  • Writing large amounts of production code
  • Optimizing prompts in isolation from system design

Desired Profile

  • Have designed AI intelligence systems inside real products
  • Think in terms of systems, state, and failure modes
  • Are fluent in RAG, memory design, and retrieval strategies
  • Care deeply about idempotency, reliability, and cost control
  • Can guide engineers through clear abstractions and contracts

Contract Details

  • Fractional / Contract
  • 10–20 hours per week
  • Opportunity for full‑time employment

How to Apply

  • Examples of production AI systems you’ve architected
  • How you think about memory, retrieval, and reasoning
  • Any diagrams, writing, or artifacts that show your system‑level thinking

Skills

RAG

Don't send a generic resume

Paste this job description into Mimi and get a resume tailored to exactly what the hiring team is looking for.

Get started free