Senior Data Scientist / ML Engineer specializing in NLP, anomaly detection, and cloud ML platforms
Remote, CAMachine Learning Engineer10 years experienceSeniorFinancial ServicesTelecommunicationsRetail
ScreenedIdentity Verified
Connect with Nandini
Nandini already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
ML/NLP practitioner who built customer-feedback topic modeling (NMF + TF-IDF) to diagnose chatbot-to-agent handovers and drove product/ops changes that reduced operational costs by 20%. Also developed LSTM-based intent recognition using Word2Vec/GloVe embeddings for semantic linking, and deployed an LSTM autoencoder for fraud anomaly detection that cut false positives by 25% while capturing 15% more fraud in A/B testing.
Experience
Machine Learning EngineerEmotionall
Lead Data Scientist - Cash securities technologyCiti
Senior Data Scientist - Consumer Products & ServicesVodafone
Data Scientist - Loyalty & AnalyticsShoppers Stop Ltd
Data Analyst - Business Transformation OfficeFirstsource Solutions Ltd
Systems EngineerInfosys Ltd
Lead Machine Learning EngineerOmdena Inc
Machine Learning EngineerOmdena Inc
Education
University of Pune (STES Shrimati Kashibai Navale College of Engineering)bachelor, Computer Science (2010)
Key Strengths
Built NLP topic modeling pipeline (NMF + TF-IDF) to analyze customer feedback and chatbot handovers
Drove measurable business impact (20% operational cost reduction) by translating insights into product/ops changes
Strong model validation practice (topic coherence, manual sampling, domain expert review, stability checks)
Designed iterative, rules-plus-sampling data cleaning process to preserve semantic meaning
Improved intent linking/search relevance using embeddings (Word2Vec/GloVe) and cosine similarity
Delivered production-grade Python workflows with orchestration, testing, logging, and scalable processing (Airflow, PySpark, BigQuery)
Fraud/anomaly detection experience with LSTM autoencoder; improved outcomes via A/B testing (25% fewer false positives, 15% more fraud captured)
Discover more candidates like Nandini
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.