MLOps Engineer
First seen by Alion on Sep 11, 2026.
Key Responsibilities :
- Build and manage MLOps/LLMOps pipelines for deployment, monitoring, evaluation, and lifecycle management.
- Productionize ML models, LLM applications, RAG pipelines, embeddings, and AI agents.
- Automate CI/CD, testing, deployment, retraining, and model/prompt versioning across the engineering stack (Angular, Django, FastAPI/LangGraph agentic service) using Jenkins.
- Implement monitoring and observability for model performance, data quality, drift, reliability, latency, and cost, including Langfuse for tracing and evaluation of the agentic AI platform.
- Build secure, scalable AI infrastructure using AWS (application infrastructure), Hetzner (GPU/model hosting infrastructure), Docker, and Terraform.
- Develop production-grade APIs, model-serving and inference systems, and data/ML pipelines.
- Support self-hosted model infrastructure on Hetzner GPU instances - model serving (e.g., vLLM or TGI for efficient LLM inference) and fine-tuning pipelines (LoRA/QLoRA), including experiment tracking and adapter versioning.
- Partner with data scientists and investment experts to operationalize AI solutions.
Qualifications & Key Skills :
- Bachelor's/Master's in Computer Science/Engineering, Data Science, or a related field.
- 5+ years of experience in MLOps, LLMOps, ML Engineering, DevOps, or Platform Engineering.
- Strong Python and software engineering skills, with CI/CD (Jenkins), cloud (AWS), Docker, and infrastructure-as-code (Terraform).
- Hands-on experience deploying, scaling, and monitoring ML models in production.
- Practical experience with LLMs, RAG, embeddings, vector databases, LLM evaluation, and AI agents.
- Experience with MLflow, Airflow, Kubeflow, or similar MLOps/LLMOps tools.
- Experience deploying and operating web applications and APIs in production (familiarity with Django and FastAPI-based services).
- Experience with GPU infrastructure and self-hosted LLM serving (vLLM, TGI, or similar); exposure to parameter-efficient fine-tuning (LoRA/QLoRA).
- Familiarity with Model Context Protocol (MCP) or similar tool-calling/agent-integration standards.
- Proficiency with Git, APIs, testing, automation, and software development best practices.
- Understanding of AI governance, security, model risk, data privacy, and responsible AI.
- Experience in financial services / investment management will be advantageous.
Skills
MLOps, LLM, Python, FastAPI, LangChain, AWS