
Fullstack LLM Application Engineer
- Genève
- CDI
- Temps-plein
Location : Geneva, Switzerland - Hybrid Work Model
Type : Permanent
Languages : Fluent English (C1+), other languages an assetAbout the Role
We are looking for an experienced Full-Stack Engineer to design, build, and deploy AI-powered chatbot applications serving mission-critical functions across trading, middle office, and back office environments. You will take end-to-end ownership of the application stack, from UI to secure backend APIs and integrations with existing enterprise systems, collaborating closely with data science and infrastructure teams.Key Responsibilities
- Chat UI Development: Create responsive React/TypeScript interfaces with SSO, semantic search, and chat patterns.
- API & Model Serving: Build FastAPI REST/gRPC endpoints, deploy LLMs using Triton, Ray Serve, or AWS SageMaker with GPU-aware autoscaling.
- Security: Implement OAuth2/OIDC, prompt validation, rate limits, audit logging, and row-level entitlements.
- Integration: Connect with Java/.NET systems, FIX, Kafka, and message queues with low latency.
- Monitoring & Deployment: Configure observability dashboards (Prometheus/Grafana), manage GitLab CI/CD, perform blue-green deployments.
- Infrastructure Automation: Contribute Terraform modules for EKS, API Gateways, Transit Gateways, and Lambda functions.
- Python 3.x, FastAPI, LangChain/Haystack, React, TypeScript
- Hugging Face, AWS Bedrock SDK, Triton Inference Server, Ray Serve, AWS SageMaker
- OAuth2, OIDC, SSO
- SQL, pgvector, Pinecone, OpenSearch
- Docker, Kubernetes/EKS, Terraform, GitLab CI
- REST, gRPC, WebSockets, SSE
- 4+ years in developing cloud-native, user-facing applications (finance/trading preferred).
- Proven deployment of LLMs or deep learning models in production.
- Strong presentation and stakeholder communication skills.
- Ability to thrive in a high-speed environment while maintaining enterprise-grade reliability.