AI engineer in financial services

We are building an AI transformation function from the ground up. We are hiring an AI Engineer to be the technical engine of this function - a hands-on builder who takes AI use cases from validated idea to production, end to end. This is a build role. You work closely with the Head of AI Transformation, who owns use case discovery and prioritization. You own the rest - product architecture, the infrastructure each use case needs, delivery estimates, and the build itself.

• Take prioritized AI use cases from proof-of-concept to production, and own the full lifecycle. Areas include customer support, sales enablement, compliance and KYC workflows, marketing content, internal operations, and analytics assistants.
• Design and build the LLM applications and agents behind these use cases - RAG pipelines, agentic workflows, tool and function calling, structured outputs, multiagent orchestration - on top of frontier model APIs (Anthropic Claude, OpenAI, and others).
• Build the fullstack around the AI: backend services, APIs, and the front-end surfaces users actually touch.
• Own deployment and operations. Deploy to the selected cloud provider, monitor, and keep things running.
• Integrate AI into the systems where work happens: CRM, ticketing and support platforms, communication tools, internal databases, and trading-adjacent back-office systems.
• Build evaluation harnesses and define quality metrics per use case. Run evals before and after changes. Instrument solutions so business impact is shown with evidence.
• Help define AI safety, data-handling, and guardrail standards at company level, not just per project, and build to them. Manage sensitive customer and financial data, prompt-injection and data-leakage risks, and design for traceability so that AI-driven actions and outputs can be logged, audited, and explained in a regulated financial services environment.
• Work closely with the Head of AI Transformation on product architecture, infrastructure, and estimates during use case evaluation: what's buildable now, what's fragile, what the real cost and timeline look like.

• Proven track record of shipping AI use cases to production - not demos, not notebooks. You can walk us through 2-3 real implementations: the business problem, what you built, what broke, and what impact it had.
• 5+ years of software engineering experience, with 2+ years building LLM-based applications.
• Strong fullstack development - backend services and APIs, plus enough front-end to ship a usable interface.
• DevOps capability - CI/CD, containers, and deployment to a major cloud provider (AWS, GCP, orAzure). You deploy and operate what you build.
• Deep working knowledge of the modern LLM application stack: prompt engineering, RAG, multi-agent/subagent orchestration, tool and function calling, and evaluation harnesses.
• Strong general engineering fundamentals: APIs, backend services, data handling, deployment, monitoring. We don't mind which languages or frameworks you prefer - we care that you understand the key components and choose pragmatically.
• Demonstrated command of the failure modes of agentic development - context and window management, model output that's plausible but wrong, loss of state across sessions, silent divergence from plan - and the engineering practices that detect and contain each.
• Experience integrating AI into existing business systems and workflows, not just standalone apps.
• Ability to operate autonomously with high ownership: ambiguous input, working software output.
• Strong communication in English - you'll explain technical trade-offs to non-technical stakeholders.
• Будет плюсом: Data pipeline experience - ETL/ELT, orchestration, warehouse or lakehouse work; Experience in fintech, brokerage, banking, or another regulated industry; Experience with voice AI, multilingual applications, or customer-facing chatbots at scale

• Ground-floor AI transformation - you're building the function, not maintaining someone else's.
• A path to lead. This is the founding AI engineering hire. As the function grows, the right person is positioned to become the lead and architect of the AI department.
• Real autonomy: short decision paths, no legacy AI infrastructure to inherit. Can be fully remote, with flexibility on location and working setup.
• A portfolio of diverse use cases across support, compliance, marketing, and operations - you won't be stuck on one product surface.