Mayank Jobanputra is an NLP-focused researcher-engineer with 11 years of experience bridging production software and academic research across Germany and India. Currently an ELLIS PhD candidate and associated member working between Universität des Saarlandes and Neuroexplicit Models of Language, Vision, Action, he focuses on language, vision and evaluation methods for natural language systems. He has industry experience as an NLP engineer at deepset and contributed backend fixes and refactors to the widely used Haystack framework, improving reliability of retrievers, embedding encoders and telemetry for real-world LLM applications. His background spans applied systems design at Aruba/HPE and hands-on research roles at IIIT Delhi and Tübingen, giving him fluency in shipping robust ML systems and advancing core evaluation techniques. Colleagues know him for pragmatic code hygiene—removing duplicated fields and tightening input types—and for shifting multimodal retrieval components toward denser, production-ready designs. Based in Saarbrücken, he combines rigorous academic inquiry with practical engineering to move research ideas into deployable NLP tools.
11 years of coding experience
4 years of employment as a software developer
Master of Arts - MA, Computational Linguistics, Master of Arts - MA, Computational Linguistics at University of Tuebingen
Bachelor’s Degree, Information and Communication Technology, 3.18 CGPA, Bachelor’s Degree, Information and Communication Technology, 3.18 CGPA at School of Engineering and Applied Science
AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.
Role in this project:
Backend Developer
Contributions:126 reviews, 20 commits, 61 PRs in 3 months
Contributions summary:Mayank primarily contributed to bug fixes and refactoring efforts within the Haystack framework. Their work included removing duplicated fields, refining input types, and updating tests related to embedding encoders and document stores. They also implemented telemetry changes and addressed model loading issues, demonstrating a focus on improving the framework's functionality and reliability. Additionally, the user refactored code to change the MultiModal retriever to be of type DenseRetriever.
Contributions:32 pushes, 2 branches in 5 years 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.