Senior AI Engineer

VerbaFlo; Gurugram District; On-site; INR 4,000,000 - 7,000,000; Full time

Job summary

VerbaFlo is seeking a Senior AI Engineer with deep RAG and embedding systems expertise. You will own end‑to‑end RAG pipelines, manage embedding infrastructure, and integrate AI features into chatbot products.

The role requires production‑oriented mindset, collaboration with product teams, and strong cloud/MLOps practices. You will work with diverse data sources, including SQL/NoSQL, PDFs, and real‑time APIs, to deliver scalable AI solutions using Python, LangChain, and vector databases.

Qualifications

• 5+ years of total engineering experience, with 3+ years in LLM or NLP engineering.
• Hands‑on experience designing and deploying RAG systems end‑to‑end.
• Deep familiarity with embedding models (OpenAI Ada, Cohere, BGE, E5) and vector databases (Pinecone, Weaviate, Chroma, pgvector).
• Strong command of Python and LLM orchestration frameworks (LangChain, LlamaIndex, Haystack).
• Experience with multiple data sources: structured, unstructured, real‑time.
• Practical knowledge of late chunking and advanced retrieval strategies.
• Cloud deployment (AWS / GCP / Azure) and containerization (Docker, Kubernetes).
• Bias for building production‑ready systems.

Responsibilities

• RAG pipeline ownership: Ideate, architect, build, and deploy end‑to‑end RAG systems from scratch through to production.
• Embedding systems: Select, evaluate, and fine‑tune embedding models; manage vector stores and optimize retrieval quality.
• Advanced chunking: Implement late chunking and other segmentation strategies to maximize context fidelity and retrieval precision.
• Multi‑source data integration: Connect and ingest from diverse sources, including SQL/NoSQL databases, PDFs, web content, Confluence, SharePoint, and real‑time APIs.
• Chatbot integration: Embed RAG and LLM components into conversational AI products using LangChain, LlamaIndex, or custom orchestration layers.
• Evaluation and quality: Own retrieval evaluation frameworks (RAGAS, triad evals) and iterate on pipelines based on precision, recall, and relevance metrics.
• Deployment and observability: Deploy and monitor LLM services on cloud infrastructure with robust logging, alerting, and MLOps practices.
• Collaboration: Partner with product and engineering teams to deliver low‑latency, reliable AI experiences at scale.

Skills

RAG

Embedding models

Python

LangChain

LlamaIndex

Haystack

MLOps

Tools

Pinecone

Weaviate

Chroma

pgvector

Docker

Kubernetes

Job description

We are looking for a Senior AI Engineer with deep specialization in Retrieval-Augmented Generation (RAG) and embedding systems. The ideal candidate has a proven track record of building end-to-end AI solutions from ideation through to production deployment and thrives at the intersection of language models, vector search, and applied NLP. This is a hands‑on, high‑ownership role. You will design and ship RAG pipelines, own embedding infrastructure, integrate AI capabilities into chatbot products, and work with diverse data sources to build systems that actually work in production.

Responsibilities

• RAG pipeline ownership: Ideate, architect, build, and deploy end-to-end RAG systems from scratch through to production.
• Embedding systems: Select, evaluate, and fine‑tune embedding models; manage vector stores and optimize retrieval quality.
• Advanced chunking: Implement late chunking and other segmentation strategies to maximize context fidelity and retrieval precision.
• Multi‑source data integration: Connect and ingest from diverse sources, including SQL/NoSQL databases, PDFs, web content, Confluence, SharePoint, and real‑time APIs.
• Chatbot integration: Embed RAG and LLM components into conversational AI products using LangChain, LlamaIndex, or custom orchestration layers.
• Evaluation and quality: Own retrieval evaluation frameworks (RAGAS, triad evals) and iterate on pipelines based on precision, recall, and relevance metrics.
• Deployment and observability: Deploy and monitor LLM services on cloud infrastructure with robust logging, alerting, and MLOps practices.
• Collaboration: Partner with product and engineering teams to deliver low‑latency, reliable AI experiences at scale.

Requirements

• 5+ years of total engineering experience, with 3+ years in LLM or NLP engineering.
• Hands‑on experience designing and deploying RAG systems end-to-end.
• Deep familiarity with embedding models (OpenAI Ada, Cohere, BGE, E5) and vector databases (Pinecone, Weaviate, Chroma, pgvector).
• Strong command of Python and LLM orchestration frameworks (LangChain, LlamaIndex, Haystack).
• Experience working with multiple data source types: structured, unstructured, and real‑time.
• Practical knowledge of late chunking and other advanced retrieval strategies.
• Familiarity with cloud deployment (AWS / GCP / Azure) and containerization (Docker, Kubernetes).
• Strong problem‑solving instincts and a bias for building things that work in production.

Good to Have

• Experience with agentic frameworks (AutoGen, CrewAI, or custom agents).
• Exposure to graph‑based RAG or knowledge graph integration.
• Open‑source contributions in the AI/ML space.