Aman Srivastava is a Lead Data Scientist based in Singapore with 11 years of experience building production-grade ML systems across search, recommendation, NLP, and risk modelling. At Pelago (Singapore Airlines) he led search, recommendation and chatbot teams driving 20% of platform bookings, shipping BERT-based semantic search, LTR/Elasticsearch ranking, FAISS de-duplication for 300K+ products and RAG-powered assistants that halved support tickets. He has deep hands-on experience optimizing pipelines to sub-200ms latency and scaling models for 100M+ user datasets from his time at Paytm and prior roles. An active open-source contributor, Aman enhanced text classification and multi-class probability prediction in the popular flairNLP project and improved multilingual NER and translation tooling for Haptik’s chatbot stack. Entrepreneurial by nature, he founded consumer apps (Notopass, Foodgreed) and open-sourced utilities like a spell-correction library, showing a knack for product-minded engineering. He pairs clear articulation and organisational rigor with a persistent hunger to refine models and systems for real-world impact.
11 years of coding experience
10 years of employment as a software developer
Master's degree Data Science, Master's degree Data Science at The University of Texas at Austin
Bachelor of Technology (BTech) Computer Science, Bachelor of Technology (BTech) Computer Science at Jaypee Institute Of Information Technology
chatbot_ner: Named Entity Recognition for chatbots.
Role in this project:
Back-end Developer & NLP Engineer
Contributions:13 reviews, 392 commits, 87 PRs in 1 year 9 months
Contributions summary:Aman focused on improving the named entity recognition (NER) functionality within the chatbot project. They addressed bugs related to fallback values and improved the handling of original text data. Further contributions include the addition of Google Translate API integration for text translation and the renaming of a translation method. Finally, the user made enhancements to a regular expression based phone number detector.
A very simple framework for state-of-the-art Natural Language Processing (NLP)
Role in this project:
ML Engineer
Contributions:8 commits, 1 PR, 9 comments in 6 days
Contributions summary:Aman primarily contributed to the `flairnlp/flair` repository by implementing and refining text classification functionalities. Their work involved adding methods to predict class probabilities for text classification models, as well as refactoring the code to improve clarity. They also incorporated and refined integration tests to ensure the correct operation of the new features. Their contributions appear focused on enhancing the model's capabilities and reliability for multi-class classification.
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.