About Us
We are an AI-focused technology company, building conversational AI, AI agents and automation solutions for healthcare, education, and enterprise and mid-market customer support. Role Overview We are moving from traditional AI to advanced, LLM-powered systems, and we want to do it cost-effectively. As an AI Engineer Intern, you will help build AI agents, conversational AI and automation for real client use cases, working closely with our senior engineering team. You will learn to choose the right approach for each problem, whether that is an LLM, a smaller open-source model or classic machine learning, balancing quality, speed and cost.

Key Responsibilities
AI Agents and LLM Applications
• Build AI agents that use tools, call APIs and complete multi-step tasks
• Develop conversational AI for chat, WhatsApp and voice channels
• Build retrieval-augmented generation (RAG) systems over client documents and data
• Write, test and refine prompts and agent workflows Cost-Effective AI
• Compare commercial APIs with open-source models for quality, speed and cost
• Run and serve open-source models efficiently using quantization and optimized inference
• Reduce costs through caching, model routing and smaller task-specific models
• Track token usage, latency and cost per conversation Models and Data
• Fine-tune small and medium models for specific tasks where it beats prompting
• Apply classic machine learning where it fits better than LLMs, such as classification, prediction and lead scoring
• Prepare, clean and manage datasets from sources such as CSV, SQL and cloud storage Evaluation and Deployment
• Build evaluation sets and measure accuracy, hallucinations and response quality
• Run experiments and A/B tests to compare approaches
• Package and deploy AI services as APIs with the engineering team
• Monitor live systems and help improve them over time Research
• Keep up with new models, agent frameworks and techniques




• Test promising tools and share practical findings with the team Essential Skills Programming
• Python: strong fundamentals, clean and readable code
• Git: version control and collaboration
• REST APIs and JSON: building and calling APIs AI and LLMs
• LLM APIs: hands-on use of OpenAI, Anthropic Claude or Google Gemini
• Prompt engineering: structured outputs, function or tool calling
• RAG basics: embeddings, chunking, vector databases such as Chroma, Qdrant or pgvector
• Agent frameworks: LangChain, LangGraph, LlamaIndex or similar
• Hugging Face: using and running open-source models such as Llama, Mistral or Qwen Machine Learning Fundamentals
• Core ML concepts: supervised learning, evaluation metrics, overfitting
• PyTorch or scikit-learn: basic hands-on use
• Data handling: pandas, NumPy and SQL Good to Have
• Running local or self-hosted models with Ollama or vLLM
• Fine-tuning with LoRA or QLoRA
• Voice AI: speech-to-text and text-to-speech tools such as Whisper
• Model Context Protocol (MCP) or other ways of connecting agents to tools
• LLM evaluation and monitoring tools such as Langfuse, LangSmith or Ragas
• FastAPI, Docker and a cloud platform such as AWS, GCP or Azure
• Interest in healthcare, education or customer support use cases Education and Experience
• BE, BTech, ME, MTech or MCA in Computer Science, AI, Data Science or a related field, completed or in final year
• Projects, a GitHub profile, hackathons or internships showing hands-on AI work; please share links

What We Offer
• Hands-on work on live AI agents and products used by real clients
• Mentorship from senior engineers
• Exposure to the full cycle: idea, build, evaluation and deployment
• Prospect for a full-time role based on performance What We Are Looking For
• Builder: ships working prototypes, not just notebooks
• Cost-aware: thinks about efficiency as well as accuracy
• Curious: experiments with new models and tools
• Research-minded: tests ideas and measures results
• Self-driven: learns quickly and works well with ambiguity

📌 Artificial Intelligence Engineer Intern (India)
🏢 AVEXHA
📍 India