AI ML Intern for Upgradation, Fine-tuning & Post-training of AI Models & Llms

About the role

This internship position places you at the center of strategic artificial intelligence advancement within the Indian Army ecosystem, where your contributions will directly influence the evolution of automated defense systems. You will engage in rigorous experimentation and development focused on enhancing the performance and reliability of large language models through structured post-training methodologies. The role demands meticulous attention to detail as you apply fine-tuning strategies that align AI capabilities with specific operational requirements. You will collaborate closely with domain experts to translate complex functional objectives into concrete model improvement pipelines. Your work will involve designing and executing validation protocols that ensure the robustness and security of deployed intelligent systems. This internship provides a unique opportunity to understand how cutting edge AI research is integrated into mission critical defense environments. You will be responsible for documenting experimental results and maintaining reproducible workflows that support long term model governance. Through this experience, you will gain deep exposure to the challenges of deploying generative technologies under strict compliance and operational standards.

Key facts

What you'll do

You will conduct comprehensive analysis of existing AI and machine learning architectures to identify optimization opportunities for enhanced performance in defense related applications. You will implement advanced fine-tuning procedures using state of the art frameworks to adapt foundational models to specialized operational contexts. You will apply supervised fine-tuning and parameter efficient methods such as Low Rank Adaptation to refine model behavior while preserving generalization capabilities. You will leverage the Hugging Face ecosystem to manage datasets, configure training workflows, and monitor model convergence across multiple experimental runs. You will construct retrieval augmented generation pipelines that integrate external knowledge sources with large language models to improve factual accuracy and context relevance. You will preprocess and curate diverse data sets, ensuring quality, consistency, and suitability for training, validation, and benchmark testing. You will execute systematic evaluation of model outputs using quantitative metrics and qualitative assessments to guide iterative improvements in system performance. You will collaborate with cross functional teams to align technical deliverables with broader project timelines and strategic objectives of the Indian Army. You will maintain detailed technical documentation that captures experimental configurations, hyperparameter choices, and observed outcomes for future reference. You will contribute to the development of scalable automation workflows that streamline model deployment and support ongoing post training activities.

Requirements

You must be pursuing or have completed a B.Tech. / B.E. or equivalent degree with a specialization in Artificial Intelligence and Data Science. You must be pursuing or have completed a B.Tech. / B.E. or equivalent degree with a specialization in Artificial Intelligence and Machine Learning. You must be pursuing or have completed a B.Tech. / B.E. or equivalent degree with a specialization in Computer Science and Engineering. You must be pursuing or have completed a B.Tech. / B.E. or equivalent degree with a specialization in Information Technology. You must be pursuing or have completed a M.Tech. / M.E degree with a specialization in Artificial Intelligence and Data Science with an emphasis in Cyber Security. You must be pursuing or have completed a M.Tech. / M.E degree with a specialization in Computer Science and Engineering. You must be pursuing or have completed a M.Tech. / M.E degree with a specialization in Information Technology. You must be enrolled in or hold a degree in any specialization demonstrating a strong foundation in quantitative methods. You must be pursuing or have completed a B.Tech. / B.E. or equivalent degree with a specialization in Artifical Intelligence and Machine Learning. You must be available to commit to the full 75 days of the internship period without interruption. You must possess relevant skills and a demonstrated interest in the project area, including hands on experience with modern AI and machine learning tools. You must be proficient in Python programming, showing the ability to write clean, efficient, and well structured code for data processing and model development. You must have working knowledge of Natural Language Processing concepts and techniques, including tokenization, embeddings, sequence modeling, and evaluation of language based tasks. You must understand the architecture and application of Large Language Models, including their capabilities, limitations, and appropriate use cases. You must have practical experience with Hugging Face libraries and tools for model training, inference, and deployment. You must be comfortable working with Generative AI systems, including data preparation, prompt design, and output validation. You must have hands on experience with PyTorch, including tensor operations, automatic differentiation, and neural network module implementation. You must demonstrate familiarity with the Transformers library for building and fine tuning models for a wide range of NLP and multimodal tasks. You must have experience in fine tuning models using techniques such as SFT to align model behavior with specific instruction sets. You must understand and apply LoRA and other parameter efficient fine tuning strategies to adapt large models with reduced computational overhead. You must have knowledge of PEFT methods that enable efficient adaptation of transformer based architectures. You must be able to implement RAG systems that combine retrieval mechanisms with language models to enhance response accuracy and timeliness. You must have a solid foundation in Data Science principles, including statistical analysis, data visualization, and experimental design. You must have strong problem solving skills and the ability to work independently on complex technical challenges while adhering to strict project deadlines.

Practical notes

Working hours: 9:30 AM To 5:30 PM

Stipend is paid after successful completion; under 75% attendance is ineligible

Police verification is required before joining

Accommodation is not provided

Personal smartphones/laptops are restricted on premises; workstations are provided

Deadline to apply: 03-Aug-2026

Internship duration: 75 Days

Start date: 17-Aug-2026

Location: Delhi, India

Credits: 15

Engagement type: Full Time

Stipend: Up to ₹75000