AI ENGINEER - REMOTO

Must-Have

- Strong Python development skills, including production-grade API and service development
- Solid knowledge of Machine Learning fundamentals, model development workflows, feature engineering, validation, and evaluation
- Hands-on experience with PyTorch and scikit-learn
- Strong understanding of Large Language Models (LLMs), Transformer architectures, tokenization, embeddings, and inference workflows
- Experience with LangChain and LangGraph for building LLM-powered and agentic applications
- Design and implementation of multi-agent and agentic AI systems, including orchestration, tool usage, state management, and guardrails
- Experience designing Retrieval-Augmented Generation (RAG) solutions, including chunking, embedding strategies, retrieval, reranking, and context management
- Experience with vector databases, preferably Qdrant, and caching/state-management technologies such as Redis
- Knowledge of fine-tuning techniques for LLMs and deep learning models, including parameter-efficient approaches such as LoRA/PEFT
- Knowledge of AI hardware stacks for development and inference, including GPUs, VRAM constraints, CUDA fundamentals, and performance-aware model deployment
- Ability to design scalable, efficient, reliable, and high-performance AI solutions, with attention to latency, throughput, cost, memory usage, and observability
- Microservices architectures
- Algorithms, data structures, and selection of the solution best suited to the context
- Docker (Dockerfiles, Docker Compose) and containerized AI workloads
- CI/CD practices for software and ML/AI systems
- Distributed systems design and integration of AI services into production architectures
- Experience in Agile environments
- Fluent English

Nice-to-Have





- Hugging Face ecosystem (Transformers, Datasets, Accelerate, PEFT)
- LLM serving and inference frameworks such as vLLM and Ray
- MLOps / LLMOps practices, including experiment tracking, model versioning, evaluation pipelines, and model lifecycle management (e.g., MLflow)
- LLM observability, tracing, and evaluation platforms such as Phoenix
- Prompt engineering and systematic evaluation of prompts, agents, and RAG pipelines
- Knowledge of model quantization, batching, caching, distillation, and other inference-optimization techniques
- Familiarity with NVIDIA GPU tooling and profiling, and with selecting appropriate hardware for training, fine-tuning, and inference workloads
- Familiarity with Kubernetes and GPU-enabled container orchestration
- Knowledge of data pipelines and orchestration tools such as Dagster
- Understanding of data quality, dataset curation, synthetic data, and responsible handling of training/evaluation data
- Familiarity with AI security topics including prompt injection, data leakage, model abuse, access control, and the NIST Cybersecurity Framework
- Experience with automated testing for AI systems, including unit, integration, regression, evaluation, and performance testing
- GitLab CI and artifact/package registries

Personal Skills

- Ability to work effectively in dynamic and evolving technical environments
- Strong problem-solving and analytical skills
- Team spirit and ability to collaborate across software, data, infrastructure, and product teams
- Commitment to continuous learning and knowledge sharing in a rapidly evolving AI ecosystem
- Proactive approach, ownership, and sense of responsibility
- Ability to balance experimentation with production quality, maintainability, security, and business constraints

📌 AI ENGINEER - REMOTO (Italia)
🏢 Dacomat
📍 Italia