Role Overview

We are looking for a skilled Generative AI Engineer with strong expertise in Python, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Agentic AI, and Machine Learning. The ideal candidate will be responsible for developing, deploying, and optimizing AI-powered applications using modern LLM frameworks, vector databases, and cloud technologies.

Key Responsibilities

- Develop AI-powered applications using Python, FastAPI, Flask, and REST APIs.

- Integrate open-source and commercial LLMs, including Llama, Qwen, Mistral, and GPT.

- Build document understanding solutions using Vision Language Models (VLMs).

- Design and implement RAG pipelines using chunking, embeddings, hybrid retrieval, and reranking techniques.

- Develop AI agents and workflows using LangGraph, LangChain, CrewAI, AutoGen, or Google ADK.

- Implement prompt engineering, function calling, structured outputs, and tool-calling workflows.

- Work with vector databases and semantic search technologies to improve information retrieval.

- Develop and optimize Machine Learning and Deep Learning models.

- Deploy and serve AI models using vLLM, Hugging Face TGI, or similar frameworks.

- Integrate AI solutions with databases, cloud platforms, and containerization technologies.

- Monitor model performance, maintain logs, and implement AI observability practices.

Required Technical Skills

Programming & Backend

- Python (Mandatory), SQL

- FastAPI, Flask, REST APIs, asynchronous programming

Generative AI & LLMs

- Open-source and commercial LLMs: Llama, Qwen, Mistral, GPT

- Vision Language Models (VLMs) for document understanding

- Prompt engineering, function calling, structured outputs, and tool calling

Agentic AI & RAG

- LangGraph, LangChain, CrewAI, AutoGen, or Google ADK

- RAG pipelines, chunking strategies, embeddings,



hybrid retrieval, and reranking

- BGE models, Sentence Transformers, and semantic search

Vector Databases & Knowledge Graphs

- Qdrant, Weaviate, Milvus, or Pinecone

- Neo4j, GraphRAG, and graph-based retrieval (preferred)

Machine Learning & Deep Learning

- Scikit-learn, supervised and unsupervised learning

- Feature engineering, feature selection, model evaluation, cross-validation, and hyperparameter tuning

- PyTorch, TensorFlow, and Hugging Face Transformers

- Neural networks, CNNs, RNNs, LSTMs, Transformers, and attention mechanisms

- Transfer learning and model optimization

Databases & Workflow Orchestration

- PostgreSQL, Redis, and MongoDB (positive to have)

- Apache Airflow (good to have), Celery

Cloud & DevOps

- AWS, Azure, or GCP

- Docker and Kubernetes

- Git, GitHub, and GitLab

- Model serving using vLLM, Hugging Face TGI, or similar technologies

Monitoring & Observability

- MLflow (preferred)

- Logging, model monitoring, and performance tracking

Preferred Qualifications

- Hands-on experience building and deploying production-ready Generative AI applications.

- Experience developing multimodal AI solutions for document processing and understanding.

- Knowledge of knowledge graphs and GraphRAG architectures.

- Experience with cloud-based AI deployments and scalable model-serving infrastructure.

- Understanding of AI application monitoring, evaluation, and optimization.

Indium is an AI services company that helps enterprises build, scale, and innovate with cutting-edge technology. Backed by the prominent private equity firm EQT, Indium has spent 26 years providing excellence in digital engineering for Fortune 500 firms. With an "AI-first by design" approach, Indium specializes in Agentic AI, Data & Analytics, Application Engineering, and Quality Engineering.

📌 Senior AI Engineer (Bengaluru)
🏢 Indium
📍 Bengaluru