Headline

Data Scientist @ UPL | ex-LeadSuared, FlyFin AI, Siemens | NIT Jamshedpur | Generative AI| Machine Learning, NLP

About

As a Data Scientist with over 2 years of experience, I specialize in designing and deploying machine learning models that address real-world challenges. My expertise spans across a range of machine learning techniques, including supervised and unsupervised learning, deep learning, and generative AI (LLMs). I have a strong background in building end-to-end ML solutions, from prototyping through production, with a focus on ensuring robust performance through feature engineering, hyperparameter tuning, and model evaluation. In the field of Natural Language Processing (NLP), I am proficient in leading NLP libraries such as PyTorch, TensorFlow, HuggingFace Transformers, and SpaCy. I leverage cutting-edge NLP techniques to extract valuable insights from both structured and unstructured data. Beyond technical expertise, I have a strong ability to translate complex data insights into actionable business strategies, working closely with cross-functional teams to drive product development. I’m passionate about staying at the forefront of AI and machine learning research, continuously exploring new methods to deliver innovative solutions that create significant business value.

Location

Bengaluru

AI stack

anthropic apis

antigravity

claude code

claude

cursor

deepmind

Work experience

Data Scientist at UPL, May 2025 - Present · 1 yr 4 mos

Data Scientist at LeadSquared, Sep 2023 - Jul 2025 · 1 yr 11 mos, Bengaluru, Karnataka, India, • Transitioned Bot Architecture: Migrated from a traditional journey-based, hardcoded bot to a text-embedding retrieval model, and ultimately to a generative AI-powered bot, enhancing bot quality and user experience. This upgrade increased customer lifetime value by 35%. • Next Text Suggestion Model: Built a finetuned T-5 model and deployed the model using FastAPI . This product accelerated agent response times, improving customer support efficiency and reducing First Response Time by 18%. • Ticket Classification Model: Trained an Anthropic Claude LLM to classify tickets into Query, Response, and Complaint (QRC) categories, streamlining ticket resolution and enhancing support operations.

Data Scientist (Product analytics) at FlyFin AI, Jun 2022 - Sep 2023 · 1 yr 4 mos, • Built product analytics dashboards, conducting feature analytics, and A/B testing. Mentored interns under the same. Performed statistical modeling, generated forecasts, and presented findings to product teams and founders. • Created a confidence meter using a KNN classifier to predict the likelihood of expenses being IRS deductible using user profile and expense detail, achieving 91% accuracy, improving financial decision-making for users. • Developed ChatGPT Tax Assistant Bot: Engineered a tax assistant bot using ChatGPT, achieving 98.3% accuracy in tax-related queries and 93.7% overall accuracy. This innovation enhanced user support and increased onboarding rates by 15%. • Built an LSTM seq2seq model for dirty payee-name cleaning for analysis and model readability purpose.

Product Analyst Intern at FlyFin AI, Dec 2021 - May 2022 · 6 mos

SDE Intern at Siemens, May 2021 - Jul 2021 · 3 mos, India, Automated Data Pipeline: Developed shell scripts on AWS S3 to automate the upload of client-shared CSV files to SAP HANA, and monitored daily upload logs for consistency. ETL Operations: Performed ETL processes on 10 CSV files prior to uploading data to ThoughtSpot, ensuring data quality and readiness for analysis. Data Visualization Report: Created a comparative report analyzing ThoughtSpot against leading data visualization tools like PowerBI, QlikSense, and Tableau, to support growth strategies.

Education

Bachelor of Technology - BTech, Computer Science and Engineering, National Institute of Technology Jamshedpur

Public links

github: https://github.com/rishabh1125