Master Thesis in Enhancing and Evaluating Natural Language to SQL Capabilities for LLM Applications

Robert Bosch Group; Reutlingen; Vor Ort; EUR 40.000 - 60.000; Vollzeit

Zusammenfassung

An innovative firm is seeking a Master Thesis candidate to enhance Natural Language to SQL capabilities for LLM applications. This exciting project focuses on bridging the gap between complex industrial data and user interaction through natural language. You will analyze existing methods, automate data processes, and implement enhancements to improve NL2SQL performance. This opportunity allows you to work at the forefront of AI technology, contributing significantly to a data visualization tool. If you are passionate about data science and eager to make an impact, this role is perfect for you.

Qualifikationen

• Master's enrollment required; good knowledge of Python and SQL.
• Experience with AI, Data Science, and web development is a plus.

Aufgaben

• Analyze NL2SQL methods and identify improvement areas.
• Automate data collection and management for training datasets.
• Implement and benchmark methods to enhance NL2SQL performance.

Kenntnisse

Python

SQL

Data Science

AI

GenAI

LLMs

Microelectronics

Web Development

Ausbildung

Master studies in Computer Science

Engineering

Microelectronics

Tools

Scripts for data extraction

Benchmarking tools

Jobbeschreibung

Master Thesis in Enhancing and Evaluating Natural Language to SQL Capabilities for LLM Applications

At Bosch, we shape the future by inventing high-quality technologies and services that spark enthusiasm and enrich people’s lives. Our promise to our associates is rock-solid: we grow together, we enjoy our work, and we inspire each other. Join in and feel the difference.

The Robert Bosch GmbH is looking forward to your application!

The increasing use of Natural Language to SQL (NL2SQL) techniques is transforming the way large language models (LLMs) help bridge the gap between complex industrial data and users, enabling domain experts to interact with data using natural language. However, challenges remain in optimizing and evaluating NL2SQL outputs, particularly for interactive AI applications and specialized domains like semiconductor data visualization. This project aims to investigate and improve NL2SQL methods to support our "Data Viz" tool.

1. During your thesis, you will analyze existing domain knowledge, advanced NL2SQL methods, and evaluation techniques, including reviewing methods such as fine-tuning, retrieval-augmented generation (RAG), and AI agent applications in NL2SQL tasks. You will identify key areas for improvement.
2. You will automate data collection, preparation, and management by developing scripts and tools to extract and preprocess training and testing datasets, ensuring data quality across various scenarios.
3. You will implement methods to enhance NL2SQL performance and retrieval accuracy, designing and benchmarking multiple approaches.
4. You will establish evaluation protocols, define metrics (e.g., accuracy, efficiency), and systematically assess improvements.
5. Finally, you will test and validate the enhanced NL2SQL system against a baseline, document methodologies, codes, processes, and results, and prepare a comprehensive final report.

Education: Master studies in Computer Science, Engineering, Microelectronics, or a comparable field.

Experience and Knowledge: Good knowledge of Python; experience with SQL, Data Science, AI, GenAI, LLMs, microelectronics, and web development is a plus.

Personality and Working Practice: Growth mindset.

Languages: Good command of English.

Start: According to prior agreement.
Duration: 6 months.

Requirement: Enrollment at university. Please attach your CV, transcript of records, examination regulations, and if applicable, a valid work and residence permit.

Diversity and inclusion are core to our culture. We welcome applications regardless of gender, age, disability, religion, ethnicity, or sexual orientation.

Further Information: Xueming Li (Functional Department)
[phone]