Backend Developer – LLM & Vector Database Integration Internship in Ahmedabad, Gandhinagar (Hybrid)

We are looking for a backend developer with hands-on experience integrating open-source LLMs (e.g., Mistral, LLaMA) and vector databases (like FAISS, Qdrant, Weaviate, Milvus) in offline or edge environments.

You will help develop core APIs and infrastructure to support embedding pipelines, retrieval-augmented generation (RAG), and LLM inference on local or mobile hardware.

Selected Intern's Day-to-day Responsibilities Include

• Design and implement backend services that interface with local LLMs and vector databases
• Develop APIs to support prompt engineering, retrieval, and inference
• Integrate embedding models (e.g., BGE, MiniLM) and manage chunking/processing pipelines
• Optimize performance for offline or constrained environments (mobile, embedded devices, etc.)
• Package and deploy models via frameworks like Ollama, llama.cpp, gguf, or on-device runtimes
• Handle local file I/O, document ingestion, metadata tagging, and indexing
  
  

About Company: Infoware is a process-driven software solutions provider specializing in bespoke software solutions. We work with several enterprises and startups and provide them with end-to-end solutions. You may visit the company website at https://www.infowareindia.com/