I am an engineer with 10+ years of experience delivering end-to-end AI, GenAI, Agentic AI, and multimodal learning solutions. I have built LLM-powered systems, automated data workflows, and deployed scalable MLOps pipelines in cloud-native environments. I thrive in fintech, banking, and regulatory reporting contexts, blending NLP, RAG, and reinforcement learning to deliver production-ready AI solutions. I enjoy collaborating with cross-functional teams to translate complex problems into practical outcomes.

Vikas Subramaniam

I am an engineer with 10+ years of experience delivering end-to-end AI, GenAI, Agentic AI, and multimodal learning solutions. I have built LLM-powered systems, automated data workflows, and deployed scalable MLOps pipelines in cloud-native environments. I thrive in fintech, banking, and regulatory reporting contexts, blending NLP, RAG, and reinforcement learning to deliver production-ready AI solutions. I enjoy collaborating with cross-functional teams to translate complex problems into practical outcomes.

Available to hire

I am an engineer with 10+ years of experience delivering end-to-end AI, GenAI, Agentic AI, and multimodal learning solutions. I have built LLM-powered systems, automated data workflows, and deployed scalable MLOps pipelines in cloud-native environments.

I thrive in fintech, banking, and regulatory reporting contexts, blending NLP, RAG, and reinforcement learning to deliver production-ready AI solutions. I enjoy collaborating with cross-functional teams to translate complex problems into practical outcomes.

See more

Experience Level

Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Intermediate
Intermediate
See more

Language

English
Fluent

Work Experience

AI Engineer at ING
February 1, 2025 - November 9, 2025
Part of Tech4Ops transformation through GenAI-powered automation, reducing manual review cycles and streamlining regulatory reporting workflows. Built an Agentic AI log parser using LangChain and Azure OpenAI for autonomous analysis and summarization of infrastructure logs, reducing triage time by 60%. Developed hybrid RAG and QLoRA models integrating structured system metrics with unstructured logs to adaptively detect novel failure patterns and reduce false positives by 35%. Enhanced LLMOps pipelines with observability, monitoring of model outputs, hallucination tracking, and drift detection to support scalable deployment with high concurrency.
ML Engineer at EY
February 1, 2025 - February 1, 2025
Delivered end-to-end ML solutions leveraging Generative AI and Agentic AI for fintech and FMCG clients. Built multimodal generative AI solutions combining text and document images to automate extraction, summarization, and verification of regulatory filings and financial documents; deployed 1,000+ pages/day processing via LLM-powered agents. Implemented Whisper-based audio transcription for financial compliance calls, enabling structured audit trails and anomaly detection. Contributed to ESG document intelligence to automate scoring and accelerate compliance workflows; designed LLMOps pipelines on Azure with CI/CD, versioning, drift detection, and observability for large-scale deployments.
Sr Data Scientist at Deutsche Bank AG
December 1, 2021 - December 1, 2021
ESG analytics support to improve data practices and automate processes. Applied ML and NLP to build sentiment analytics for investors from 500k social media posts/month, informing quicker portfolio decisions. Built an ESG analytics toolkit for client advisory using ensemble models to assess KPI-driven risk and accelerate compliance workflows.
Sr Data Scientist at Barclays
April 1, 2021 - April 1, 2021
Developed time-series forecasting for interest rate exposures and explored LSTM/VAR approaches. Built mortgage risk underwriting models using XGBoost with notable lift over prior generations and improved customer paid rates via Digital Affinity ML models. Implemented backend modules with Python/Flask and designed Azure-based data pipelines to reduce manual intervention and improve throughput.
Data Scientist at Morgan Stanley
May 1, 2019 - May 1, 2019
Part of MSAS trading group; developed multi-factor models for trading risk, computed VaR and stressed VaR for emerging markets. Built regression-based econometric frameworks and data pipelines to support risk assessment and reporting.
Data Analyst at Crisil
February 1, 2017 - February 1, 2017
Worked on probability of default modeling for unsecured loan portfolios and stress testing to assess macroeconomic sensitivity. Built early risk analytics to support portfolio decisions.
Business Analyst at Evalueserve
February 1, 2016 - February 1, 2016
Index reporting and analytics (LVCI), developing research-backed index methodologies, and supporting corporate actions validations. Conducted hate speech detection research and authored whitepapers on multimodal LLMs for ESG scoring and Agentic AI.

Education

MSc Economics at Madras School of Economics
August 1, 2011 - June 1, 2013
BSc Physical Science at University of Delhi
July 1, 2008 - July 1, 2011

Qualifications

Add your qualifications or awards here.

Industry Experience

Financial Services, Professional Services, Software & Internet