Solutions Architect AI Systems
Movadex · UA ·
- Category
- Architecture
Movadex · UA ·
Movadex Labs is looking for a highly technical Solutions Architect to help design and deploy custom AI systems for enterprise clients. This is not a generic consulting role. You will work at the intersection of business workflows, AI infrastructure, data architecture, agentic systems, and production engineering.
You will work closely with our Business Analyst, product/enterprise design lead, and engineering team to translate messy enterprise needs into clear technical architectures, sandbox environments, implementation plans, and deployable AI solutions.
The ideal person understands how to move from “the client has a workflow problem” to “here is the architecture, data flow, model/tooling layer, security approach, integration plan, and technical execution roadmap.”
What you’ll do
Work with the BA to understand client workflows, internal systems, data sources, approval flows, users, edge cases, and business constraints.
Design technical discovery frameworks and AI sandboxes for enterprise clients, including data ingestion, knowledge layers, vector search/RAG, workflow orchestration, evaluation loops, permissions, observability, and deployment options.
Architect custom agentic systems that can operate across internal documents, structured data, APIs, tools, and human-in-the-loop processes.
Define the right AI infrastructure for each use case: LLM providers, model routing, embeddings, vector databases, orchestration frameworks, retrieval pipelines, eval systems, monitoring, guardrails, and hosting environment.
Create implementation blueprints for engineering teams, including system diagrams, integration maps, API requirements, data requirements, deployment approach, security considerations, and phased delivery plans.
Support enterprise pilots from technical discovery through sandbox build, prototype, production deployment, and post-launch optimization.
Help evaluate whether a workflow should be solved with RAG, fine-tuning, workflow automation, agent orchestration, rules-based logic, traditional software, or a combination of approaches.
Partner with product/design to make sure the AI system is not just technically functional, but actually usable, adoptable, and measurable inside the client’s organization.
Collaborate with engineers during implementation, review architecture decisions, unblock technical issues, and ensure the final system matches the intended design.
Help build reusable internal assets, including reference architectures, agent templates, sandbox environments, evaluation patterns, deployment playbooks, and technical documentation.
What we’re looking for
Strong experience designing production software systems, ideally in enterprise, AI, data, SaaS, automation, or cloud environments.
Deep understanding of modern AI infrastructure: LLM APIs, RAG, embeddings, vector databases, agent frameworks, tool calling, evals, prompt orchestration, model routing, guardrails, observability, and deployment tradeoffs.
Ability to work with messy, ambiguous enterprise requirements and turn them into clear technical plans.
Strong understanding of backend systems, APIs, databases, cloud architecture, authentication, permissions, security, and integration patterns.
Comfortable designing sandbox environments and proof-of-concept systems that can later scale into production.
Ability to communicate clearly with both technical and non-technical stakeholders.
Strong documentation skills: architecture diagrams, technical specs, implementation plans, and engineering handoff materials.
Practical, product-minded thinking. You should know when to use advanced AI infrastructure and when a simpler workflow automation or software solution is the better answer.
Experience with AWS, Azure, or GCP is strongly preferred.
Experience with Python, TypeScript/Node.js, FastAPI, LangChain/LangGraph, LlamaIndex, OpenAI/Anthropic APIs, vector DBs, or similar tools is a strong plus.
You are a strong fit if
You can sit in a client discovery session, understand the business process, ask sharp technical questions, and quickly see what kind of AI system should be built.
You are not just an engineer and not just a consultant. You can bridge strategy, architecture, and implementation.
You understand that enterprise AI is not only about prompts or models. It is about workflows, data readiness, integrations, permissions, evals, adoption, and measurable ROI.
You can design something that engineers can actually build.