Platform Architect
capital.com5 days ago
Limassol, Cyprus or Warsaw, PolandSenior
Responsibilities
- Define and own the reference architecture and all architectural decisions for the governed enterprise AI platform.
- Evaluate models, tools, and frameworks and manage enterprise technical relationships with Anthropic and OpenAI.
- Build and operate AI workload infrastructure on AWS and GCP, including model services, storage, identity, and monitoring.
- Deploy and securely configure n8n automation workflows connecting LLMs with Jira, Confluence, Slack, Google Workspace, HR systems, and proprietary data.
- Design MCP servers and gateways connecting AI agents with internal services.
- Build Python integrations, RAG pipelines, agent frameworks, prompt libraries, and internal AI assistants.
- Implement AI data privacy, security, observability, cost allocation, policy controls, and audit trails for regulated environments.
- Create AI adoption standards, onboarding materials, workshops, documentation, and direct support for Technology teams.
- Report platform status, risks, and roadmap to Technology leadership.
Requirements
- 8+ years of engineering experience, including at least 2 years building LLM-based systems in production.
- Production-grade Python development experience and hands-on experience with Anthropic and/or OpenAI APIs, including function calling, tool use, streaming, RAG, prompt management, token optimization, and model versioning.
- Hands-on experience designing agentic platforms, MCP servers, MCP gateways, and flows between AI agents and internal services.
- Experience deploying, securely configuring, and maintaining n8n workflows at organizational scale.
- Production experience with AWS AI services and infrastructure, including Bedrock, SageMaker, Lambda, S3, IAM, and CloudWatch.
- Production experience with GCP AI services and infrastructure, including Vertex AI, Cloud Functions, Cloud Storage, IAM, and Cloud Monitoring.
- Hands-on DevOps capability and experience deploying and operating built systems.
- Strong end-to-end architecture, system design, trade-off documentation, and stakeholder communication skills.
- Experience with secrets and credential management, key rotation, least-privilege IAM, data classification, prompt and output controls, and audit trails.
- Knowledge of LLM threats such as prompt injection, model-output data exfiltration, insecure tool and agent use, and model supply-chain risk, with working knowledge of OWASP LLM Top 10 or NIST AI RMF.
- Ability to operate as the first person in a function and deliver independently without an existing team.
- Preferred qualifications include fintech or financial-services experience, DORA or ISO 27001:2022 knowledge, familiarity with the EU AI Act, NIST AI RMF or ISO/IEC 42001, LangChain or LlamaIndex, LLM guardrail and DLP tooling, internal developer platforms or AI centers of excellence, and relevant AWS or Google professional certifications.
Benefits
- The role is the first person in the AI platform function and offers direct ownership of architecture, delivery, governance, and roadmap.
- Includes internal enablement through workshops, documentation, onboarding materials, and direct team support.
Tech Stack
Categories
About capital.com
Capital.com is a retail trading platform that lets individuals trade CFDs on shares, indices, forex, commodities, and crypto via web and mobile. Founded in 2016 and privately held, it is headquartered in Limassol, Cyprus, with regulated entities in the UK and EU; it earns revenue from spreads, overnight financing, and other brokerage fees. The platform also offers education resources and real-time market insights.