AI Engineer (RAG Specialist)
NexOne
AI Engineer (RAG Specialist)
We are looking for a skilled
AI Engineer
specializing in
Retrieval-Augmented Generation (RAG)
to join our team. Your primary focus will be bridging the gap between static LLMs and dynamic, proprietary data. You won't just be "calling an API"; you will be architecting the entire data lifecycle-from ingestion and chunking strategies to advanced retrieval and response synthesis. The ideal candidate understands that the secret to a great RAG system isn't just the LLM, but the quality of the retrieval and the nuances of the vector database.
US Citizenship Required
Key Responsibilities
•
Pipeline Architecture:
Design and deploy end-to-end RAG pipelines using frameworks like
LangChain ,
LlamaIndex , or
Haystack .
•
Data Engineering:
Develop robust ETL processes to ingest unstructured data (PDFs, docs, web scrapes) into high-performance vector stores.
•
Retrieval Optimization:
Implement and tune advanced retrieval techniques, including
Hybrid Search
(keyword + semantic),
Re-ranking
(Cross-Encoders), and
Parent-Document Retrieval .
•
Vector Database Management:
Manage and scale vector databases such as
Pinecone, Weaviate, Milvus, or Chroma .
•
Evaluation & Benchmarking:
Establish rigorous evaluation frameworks (e.g.,
RAGAS ,
TruLens ) to measure faithfulness, relevancy, and hit rates.
•
Performance Tuning:
Optimize embedding models and prompt engineering to reduce latency and "hallucinations."
Technical Qualifications
•
Language Proficiency:
Advanced
Python
(preferred) or TypeScript.
•
LLM Expertise:
Hands-on experience with OpenAI GPT-4, Anthropic Claude, or open-source models like Llama 3 via
Ollama
or
vLLM .
•
Vector Expertise:
Deep understanding of embeddings, similarity metrics (Cosine, Euclidean), and indexing strategies (HNSW, IVF).
•
NLP Fundamentals:
Familiarity with tokenization, context windows, and attention mechanisms.
•
Cloud/DevOps:
Experience deploying AI applications on
AWS, GCP, or Azure
using Docker/Kubernetes.
Preferred Skills
• Experience with
Agentic RAG
(Multi-step reasoning and tool-use).
• Knowledge of
Graph Databases
(Neo4j) for GraphRAG implementations.
• Contributions to open-source AI projects.
• Background in traditional Information Retrieval (Elasticsearch/Solr).