Lavori attuali relativi a AI Engineer - Napoli, Campania - Aieng

Aieng
  is a
high-growth innovative startup
focused on delivering next-generation engineering solutions. We bridge the gap between ambitious vision and technical reality by investing heavily in research, development, and top-tier talent. Currently in a dynamic phase of expansion, we pride ourselves on our agility and our ability to navigate complex industrial challenges. At Aieng, we don't just follow industry trends—we aim to set them. Join us as we build the infrastructure of tomorrow.

Your Role

As an
AI Engineer (Inference & RAG Architect)
, you will design, optimize, and operate
local, production-grade LLM systems
, owning the full lifecycle from low-level inference performance to high-level semantic memory and agent orchestration.

Key Responsibilities

• Architect and optimize
  high-throughput LLM inference pipelines
• Design and implement
  enterprise-grade RAG systems
• Benchmark, validate, and fine-tune open-source models for domain-specific workloads
• Build
  agentic AI systems
  with deterministic, auditable behavior
• Ensure scalability, observability, and reliability of AI systems in production

Technical Skills (Hard Skills)

LLM Inference & Systems Optimization

• Advanced configuration and tuning of
  vLLM
  ,
  Ollama
  , and
  TGI
• Deep understanding of
  PagedAttention
  , continuous batching, and KV-cache optimization
• Model
  quantization
  techniques (INT8, INT4, GPTQ, AWQ, GGUF)
• GPU scheduling
  , VRAM optimization, multi-GPU and multi-node inference
• CUDA-aware performance tuning (conceptual and practical)
• Deployment of LLMs in
  on-prem, edge, and air-gapped environments

Retrieval-Augmented Generation (RAG) & Knowledge Systems

• Design of
  multi-stage RAG pipelines
• Integration with
  vector databases
  (Qdrant, Weaviate, FAISS)
• Hybrid retrieval strategies (dense, sparse, BM25)
• Re-ranking
  using cross-encoders and LLM-based rankers
• Metadata-driven access control and document-level security
• Chunking, embedding strategy design, and context window optimization

Model Lifecycle & Evaluation

• Evaluation and benchmarking of models such as
  Llama 3, Mistral, Phi, Mixtral
• Domain adaptation via
  LoRA / QLoRA
• Prompt and system prompt engineering with reproducibility guarantees
• Offline and online evaluation frameworks (faithfulness, groundedness, latency, cost)
• Versioning and rollback strategies for models and prompts

Agentic Architectures & Orchestration

• Design of
  agent-based systems
  with tool use, memory, and planning
• Development using
  Semantic Kernel
  , LangGraph, or custom agent frameworks
• Deterministic execution, guardrails, and fallback strategies
• Implementation in
  Python and C#

MLOps, DevOps & Observability

• Containerization with
  Docker
  and orchestration via
  Kubernetes
• CI/CD for AI systems
• Monitoring of latency, throughput, hallucination rates, and failures
• Logging, tracing, and observability for LLM pipelines
• Infrastructure-as-Code (Terraform or equivalent)

Soft Skills

• Strong
  system-level thinking
  and architectural mindset
• Obsession for performance, reliability, and correctness
• Ability to translate business requirements into technical architectures
• Clear communicator in cross-functional, high-complexity environments
• Ownership mentality and engineering rigor

Experience & Education

• 3+ years of experience in
  AI Engineering, ML Systems, or Platform Engineering
• Strong academic background in
  Computer Science, Engineering, or related fields
• Proven experience deploying
  self-hosted LLMs
  in production
• Exposure to enterprise constraints (security, compliance, scalability)