Skill set
Additional skills (6)
Karthikeyan in a few words
Experienced Data Scientist with a demonstrated history of working in the Information Technology and Service Industries. Skilled in Machine Learning, Statistics, NLP, Deep Learning, LLM, Computer Vision and Big Data Analytics, Elastic Search, Kibana, Microsoft Excel, SQL, Tableau, Power BI, Looker, QlikSense, OBIEE, Strong Business Development Professional with a PGP in Data Science from International School of Engineering. Certified Google cloud Professional Machine Learning Engineer, AWS - Associate Solutions Architect.
Experience
- SPS Pvt. LtdMachine Learning EngineerMay 2018 - May 2019 (1 year)Tamil Nadu, India• Identified bottlenecks to reduce customers' time navigating support channels by 6 minutes through data visualization.• Worked with product team to build production recommendation engine in Python, improving time on page for users and resulting in $325K in incremental annual revenue.• Led a team of 3 data scientist to model the ordering process 5 unique ways, reported results, and made recommendations to increase order output by 9%• Coordinated with the product and marketing teams to determine what kind of client interactions resulted in maximized service opt-ins, increasing conversions by 18%• Developed and implemented an NLP-based system to detect and correct errors in text, resulting in a 25% reduction in customer complaints related to text errors.• Collaborated with a team of data scientists to develop and maintain an NLP model to improve accuracy and performance, resulting in a 15% increase in precision and recall metrics.• Designed and developed an NLP-based application to automate text-based tasks, reducing manual processing timeby 50% and increasing team productivity by 30%.
- INCEDO INCSenior/Lead Data ScientistMay 2019 - Today (6 years)New Jersey, USA• Led a team of 5 data scientist and 3 data engineers in developing Churn prediction models that improved to reduce Churn by 18%.• Implemented Large Language Models (LLM) for customer query understanding and document search optimization. Enhanced model efficiency by integrating LLM with existing NLP frameworks, resulting in a 20% improvement in retrieval accuracy.• Led projects related to network failure prediction, customer churn analysis, and search engine optimization.• Developed a predictive model to analyze and predict potential network failures in 5G networks, reducing downtime by 15%.• Built and deployed a customer churn prediction model for an investment bank, decreasing churn by 18%.• Created a similarity search engine using Elasticsearch to improve data retrieval speed by 30%, optimizing access to financial documents.• Constructed a real-time reporting infrastructure that standardized metrics across the company, saving 210 monthly hours of manual reporting.• Worked with product and marketing teams to identify which customer interactions during free trial maximized the likelihood of conversions, resulting in a conversion rate increase of 20%• Focused on customer retention by building customer segment models and recommended segment attention that increased retention by 21%.• Deployed a recommendation engine to production to conditionally recommend other menu items based on past order history, increasing average order size by 7%• Led a team of 3 data scientist to model the ordering process 5 unique ways, reported results, and made recommendations to increase order output by 9%• Collaborated with product managers to perform cohort analysis that identified an opportunity to reduce pricing by 21% for a segment of users to boost yearly revenue by $560,000.• Constructed operational reporting in Tableau to improve scheduling contractors, saving $90,000 in the annual budget.
- INSOFEIntern Data ScienceJanuary 2018 - March 2018 (2 months)• Developed complex data structures to support the generation of business insights and strategy.• Worked on the thorough Metadata and Data analysis to make sure that all the sensitive data has been identified.• Collaborated with clients to understand their operational and strategic needs in terms of data analysis.• Used R to parallelize pre-existing code to make data analysis processes quicker.• Generated portfolio reports daily for review by financial advisers - Conducted research into individual securities to assist investing strategies.• Utilized investment software to analyse portfolio performance and project future returns based on underlying investments and various risk factors.
Recommendations
Education
- B.E., Electronics, and InstrumentationMeenakshi college of engineering Anna university2016B.E., Electronics, and Instrumentation
- PGP, Data ScienceInternational School of Engineering, Carnegie Mellon University2018PGP, Data Science