About the job Mission: AI Cloud Engineer (AWS) | MLOps – Tenjah Mission type: Freelance / One-off engagement (part-time or full-time, to be defined) Location: Remote (Tunisia or international) Organization: B2H AI Solutions – Tenjah Project Estimated duration: To be defined based on scope (short-term mission with possibility of extension)
Context Tenjah is an EdTech platform built on a RAG (Retrieval-Augmented Generation) architecture, currently deployed in production and used by more than 200 active users across 16 school subjects. MVP version is available at the following link: https://bactunis.vercel.app/
As Tenjah scales up, we are looking for an AWS Cloud Engineer specialized in MLOps to structure, secure, and optimize the infrastructure supporting the platform's RAG pipelines and LLM services.
Mission objectives Design and deploy a scalable AWS infrastructure to host Tenjah's services (API, vector database, RAG pipelines). Implement MLOps practices for the model and inference pipeline lifecycle (versioning, monitoring, continuous deployment). Industrialize the current deployment to reduce reliance on ad hoc solutions and improve production reliability. Optimize infrastructure costs related to LLM calls and vector storage. Set up robust monitoring and observability (logs, metrics, alerting). Core responsibilities Architect and deploy the AWS infrastructure (EC2/ECS/Lambda, S3, RDS or equivalent, VPC, IAM). Set up CI/CD pipelines for deploying backend services and AI components (FastAPI, LangChain). Design MLOps pipelines: automated training/fine-tuning where relevant, model version management, regression testing on RAG responses. Integrate and optimize a vector database (Pinecone, Weaviate, or pgvector) for the RAG pipeline. Set up monitoring (CloudWatch, Grafana/Prometheus or equivalent) and alerting for critical services. Secure the infrastructure (secrets management, least-privilege IAM, encryption of user data). Document the architecture and deployment procedures to ensure project continuity. Collaborate with the founding team to prioritize technical evolutions according to the product roadmap. Tech stack Cloud: AWS (EC2, ECS/Fargate, Lambda, S3, RDS, CloudWatch, IAM, VPC) Backend: FastAPI, Python AI / RAG: LangChain, vector databases (Pinecone/Weaviate/pgvector), LLM APIs (Gemini, Groq, OpenRouter) MLOps: Docker, CI/CD (GitHub Actions or equivalent), monitoring (CloudWatch/Grafana/Prometheus) Frontend (context, not required for this mission): React/TypeScript Profile sought Proven experience in AWS cloud engineering, ideally with a relevant certification (AWS Solutions Architect, AWS Certified DevOps Engineer or equivalent). Hands-on MLOps experience: deploying and monitoring LLM/RAG pipelines in production. Strong command of Docker and CI/CD practices. Knowledge of RAG architectures and vector databases. Awareness of cloud cost optimization, particularly for LLM inference workloads. Autonomy, rigor, and the ability to document work clearly. A plus: experience in an EdTech context or a fast-growing startup. Expected deliverables Operational, documented AWS infrastructure. Functional CI/CD pipeline for backend and AI deployments. Monitoring and alerting dashboard in place. Complete technical documentation (architecture, deployment procedures, incident runbook). How to apply Please send a CV, portfolio (GitHub or equivalent), and a brief note describing relevant experience deploying AI/MLOps infrastructure on AWS.