Artificial Intelligence Engineer

Factspan Analytics; Bengaluru; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

Factspan Analytics seeks an experienced LLM Engineer to build GenAI applications, focusing on Retrieval Augmented Generation, vector stores, and secure deployment of LLMs. You will work with offshore and onsite teams to deliver scalable AI solutions.

You will design RAG pipelines, integrate vector databases, and develop secure features using Python, FastAPI, Docker, and Kubernetes in an enterprise-ready environment.

Qualifications

• 5+ years total experience in AI/ML/NLP.
• At least 2 years hands-on LLM engineering experience.
• Bachelor's or Master's in CS, DS, or related fields.
• Experience designing GenAI applications and RAG workflows.

Responsibilities

• Design and develop LLM-driven applications for customer support copilots, merchandising intelligence, document summarization, and internal productivity tools.
• Build scalable and secure RAG pipelines using LangChain, LlamaIndex, or similar frameworks.
• Integrate vector databases for semantic search.
• Work with LLM APIs and handle context engineering, chaining, and tool invocation.
• Develop APIs and backend services using Python, FastAPI, and manage deployment workflows with Docker/Kubernetes.
• Ensure logging, rate limiting, and data compliance for enterprise readiness.
• Collaborate with DevOps and data engineers to support CI/CD and monitor LLM pipeline performance.
• Work in agile delivery mode with daily scrums and backlog grooming.

Skills

Python

GenAI frameworks

LLM APIs

Prompt engineering

Vector databases

RAG pipelines

MLOps/DevOps

Team collaboration

Education

Bachelor's or Master's in CS/DS

Tools

LangChain

LlamaIndex

Haystack

Vector stores

Job description

Role Overview

We are looking for a LLM Engineer with strong hands-on expertise in building GenAI applications, especially around Retrieval Augmented Generation (RAG), vector stores, prompt engineering, and secure deployment of LLMs. As part of the offshore team, you will collaborate closely with onsite leads and U.S. client stakeholders to design and implement intelligent AI solutions at scale. .

Key Responsibilities:

• Design and develop LLM-driven applications for use cases such as customer support copilots, merchandising intelligence, document summarization, and internal productivity tools.
• Build scalable and secure RAG pipelines using LangChain, LlamaIndex, or similar frameworks.
• Integrate vector databases like FAISS, Weaviate, Milvus, or Pinecone for semantic search.
• Work with LLM APIs (OpenAI, Azure OpenAI, Mistral, Claude) and handle context engineering, chaining, and tool invocation.
• Develop APIs and backend services using Python, FastAPI, and manage deployment workflows using Docker/Kubernetes.
• Ensure clean logging, rate limiting, and data compliance for enterprise readiness.
• Collaborate with DevOps and data engineers to support CI/CD and monitor LLM pipeline performance.
• Work in agile delivery mode with daily scrums, sprint planning, and collaborative backlog grooming.

Key Skills:

• Strong proficiency in Python and one or more GenAI frameworks (LangChain, LlamaIndex, or Haystack).
• Solid experience working with LLM APIs and prompt engineering.
• Hands-on exposure to vector databases and RAG implementation.
• Understanding of token usage, function calling, and context window constraints in LLMs.
• Familiarity with MLOps/DevOps deployment in cloud or on-prem environments (preferably Azure).
• Ability to work collaboratively with cross-location teams and communicate effectively with onsite stakeholders.

Good to have Skills:

• Experience in building AI copilots or internal productivity tools.
• Familiarity with front-end tools like Streamlit or Gradio.
• Exposure to unstructured document processing or OCR.
• Retail domain knowledge is a plus.

Required Qualifications:

• 5+ years total experience in AI/ML/NLP.
• At least 2 year of hands-on LLM engineering experience.
• Bachelors or Masters in Computer Science, Data Science, or related fields.