Sr. Machine Learning Engineer/ AI Engineer

The Company

We are helping our client, a fast-growing SaaS product company serving thousands of businesses globally, scale up its AI division at the core of a next-generation FinTech platform. The company processes high-volume financial operations where intelligent automation solves real complexity workflows. With proven product traction, strong backing, and ambitious AI expansion plans, this combines production scale with genuine technical challenges and the freedom to establish foundational architecture.

The Role

You'll join a focused team of ML/ AI engineers building end-to-end LLM systems that automate complex financial workflows at scale. This is a high-ownership role where you'll have real say in what gets built, how it's architected, and how it ships to production—working on genuinely complex problems at the intersection of agents, web- and vector-based RAG, and knowledge graphs using modern tooling like LangGraph, LangChain, and Neo4j.

What You'll Do

Design and build distributed agentic systems using LangGraph—multi-agent orchestrators, parallel swarms, supervisor-worker topologies, stateful workflows

Architect knowledge graph infrastructure and semantic layers enabling agents to reason about business relationships, dependencies, and workflows

Build production graph and semantic-analysis pipelines - entity resolution, relation extraction, graph embeddings, community detection—powering anomaly signals and intelligent automation

Deploy ML services with FastAPI + Celery + Redis (I/O-bound) and Ray Serve on AKS (compute-bound); build evaluation frameworks and observability with LangFuse

Contribute to AI platform architecture—skills hub, MCP infrastructure, Python SDK, internal AI-as-a-Service tooling

Collaborate closely with product and engineering to ship AI features with high accuracy

What You Bring

3+ years production ML experience

4+ years with strong Python

Hands-on experience building LLM-powered applications (agents, RAG, prompt engineering, tool use)

Production experience with at least one agent framework (LangGraph, LangChain, or similar), including multi-agent / distributed agent designs

Solid foundation in classical ML (classification, clustering, anomaly detection) and graph / semantic-analysis techniques (a plus)

Knowledge graphs in production: ontology design, property graphs, GraphRAG retrieval, semantic search

Classical ML (classification, clustering, anomaly detection) and graph techniques (entity resolution, graph embeddings, GNN methods, community detection)

Familiarity with async processing patterns (Celery, Redis, or similar)

Production deployment (FastAPI, Ray Serve) and async patterns (Celery, Redis)

Russian and English proficiency

Nice to Have:

Knowledge graph frameworks (Cognee, GraphRAG, LlamaIndex KG) or Neo4j/NetworkX experience

Agent platform/SDK design for internal teams; MCP servers or plugin infrastructure

Azure services (OpenAI, Container Apps, AKS), FinTech domain, ML observability (LangFuse, LangSmith, W&B), LLM evaluation frameworks

Company’s tech stack:

- Languages: Python (strict typing)

- Agent Frameworks: LangGraph, LangChain

- Knowledge & Retrieval: Custom ontologies, GraphRAG, NetworkX, pgvector, Azure Document Intelligence

- Deployment: FastAPI + Celery + Redis on Azure Container Apps (primary); Ray Serve on AKS for compute-bound workloads

- Observability: LangFuse (self-hosted)

- LLM Gateway: OpenRouter, Azure OpenAI

The Company Offers

A competitive compensation package with regular reviews and additional perks, as well as remote-friendly work conditions. You'll work in a team of experienced professionals building foundational AI infrastructure—agentic platform, knowledge graphs, internal AI-as-a-Service—from the ground up, tackling complex problems at the intersection of multi-agent systems, semantic reasoning, and financial workflows with direct impact on features used by thousands of businesses.

Direct employment with support on work-permit/visa paperwork if needed.