Job Description

IRIS.AI is an ML driven software that R&D personnel all over the world throughout their work. We are building an Agentic Systems that can read and understand unstructured complex text and give you the full picture of the information inside in a human understandable way. Initially starting from research documentation we gradually expanded to complex tech specifications, schematics and images to give most value to our users and customers.

Our AI facilitates our clients’ processes with the help of:

• Agentic Workflows

• Rag-as-a-Service

• AI Chatbots

• Advanced data extraction and indexing

• Utilization of LLMs

Research Projects for this internship program:

1. “ Validation and improvement of context grounding metric”

   This project focuses on evaluating and improving context grounding metrics for Retrieval-Augmented Generation (RAG) systems, ensuring that model outputs rely on the context from provided documents. The work will analyze how current metrics, including Iris.ai ’s ConSens , perform in real-world scenarios with long answers and complex contexts. The goal is to propose and implement improvements to enhance metric validity, efficiency, and robustness, culminating in a detailed report and upgraded Python implementation.

2. “Knowledge Distillation for Efficient LLM Systems“
   This project focuses on Knowledge Distillation (KD) for efficient LLM systems at Iris.ai . The main goals are to identify effective KD strategies, develop an internal framework for model distillation, and validate distilled models in Agentic AI and RAG systems. The project will involve a literature review of KD and Small Language Models (SLMs), building an experimental setup for KD, and implementing and testing distilled models within Iris.ai ‘s codebase. Key deliverables include a comparative analysis of KD techniques, a custom prototype for KD framework modules, and an evaluation report comparing teacher and student models.

3. “Developing a framework for automatic prompt optimization”

   This project aims to develop an adaptive framework for automatic prompt optimization to improve Iris.ai’s use of large language models. Key tasks include reviewing methods, evaluating frameworks, designing a custom system, and testing in RAG scenarios. Deliverables include a comparative analysis, a custom framework, and guidelines for creating consistent, efficient prompts.

4. “ GraphRAG for content overview summarization”

   This project explores adapting the GraphRAG approach to generate high-level content summaries across large document corpora. The goal is to move beyond focused, dense outputs and enable community-based overviews that support open-ended queries and RAG indexing. Research questions address tailoring summaries to domains, strategy selection at each step, and evaluation methods. Deliverables include implementing GraphRAG summarization, benchmarking against standard map-reduction methods, and defining criteria for when to apply the approach..

5. “Evaluating AI generated summaries”

   This research project focuses on developing and refining methods to evaluate the quality of AI-generated document summaries produced by large language models (LLMs). The goal is to identify, implement, and validate metrics for assessing the factuality, completeness, coverage, and readability of summaries, addressing challenges with long and complex documents. Deliverables include a review of existing evaluation methods, Python implementations of metrics, and an analysis of their performance across various datasets.

Internship Details:

• 3-6 months long

• full-time (40 hours a week)

• fully Remote

• insightful, deeply practical and progressive internship program

• it could be Master Thesis project as well

Requirements:

• Being a Bachelor/Master student in а computer science major

• Seeking a career with Machine Learning/NLP

• Interest in and having some knowledge of NLP

• Some experience in Python development is mandatory

• Some knowledge and experience in Machine Learning

• Interest in Iris.ai’s research projects and domain

• Ability to work full-time for the period of the internship

• Located within the European Timezone (+/-2 hours of CET)

If you think you match the profile, please send us your CV in English. We appreciate everyone’s effort, however we will only contact the candidates who best meet the above requirements.

Job Tags

Similar Jobs