Our client, a leader in telecommunication technologies, is looking for a senior AI Engineer to lead IVR and Conversational AI initiatives and design scalable LLM and NLP solutions for high-volume customer interactions.
This role requires onsite presence 3 days per week in Charlotte, NC. Qualified candidates must be able work without need for sponsorship.
Key Responsibilities
- Architect and develop AI-driven IVR and conversational systems using Python
- Build NLP models for text classification and clustering to support intent detection and call routing
- Perform data analysis using SQL on large-scale IVR and interaction datasets
- Design and deploy LLM workflows (prompt engineering, RAG, evaluation) for voice and chat use cases
- Lead development of agentic AI workflows for multi-step customer interactions
- Establish LLMOps (Large Language Model Operations) best practices, including monitoring, evaluation, and lifecycle management
- Partner with IVR platform, speech engineering, and product teams
- Mentor offshore and junior AI engineers
Required Skills & Experience
- 5+ years of experience in AI / ML
- Very strong proficiency in Python
- Proven expertise in NLP, including:
- Text classification
- Clustering
- Strong experience in data analysis using SQL
- Hands-on experience with LLMs:
- Prompt engineering
- Evaluation frameworks
- Retrieval-Augmented Generation (RAG)
Nice to Have:
- Experience with Agentic AI and multi-agent orchestration
- Experience with LLMOps / MLOps in production environments
- Familiarity with IVR platforms and speech technologies such as ASR (Automatic Speech Recognition), NLU (Natural Language Understanding), and TTS (Text-to-Speech)
- Python / ML: NumPy, Pandas, Scikit-learn, XGBoost
- NLP: spaCy, Hugging Face Transformers, NLTK, Sentence-Transformers
- LLM Frameworks: LangChain, LlamaIndex, OpenAI SDK
- Agentic AI: LangGraph, AutoGen, CrewAI
- Data & SQL: Oracle, MySQL, SQLAlchemy
- Vector Databases: FAISS, Pinecone, Weaviate, Qdrant
- LLMOps / MLOps (Good to Have): MLflow, Weights & Biases, BentoML, Airflow