Monil Shah

About Monil Shah

I’m an Applied AI/ML Engineer with 10+ years of experience building quantitative and data-driven systems in financial services. My work focuses on designing and deploying production AI solutions — including GenAI, RAG-based LLM agents, explainability platforms, and automated model monitoring frameworks — that make complex risk and modeling workflows more transparent, reliable, and scalable.I build agentic AI systems that enable natural-language model interrogation, automated reporting, and continuous performance evaluation, combining LLMs with structured model outputs, APIs, and governance controls. I also have a strong foundation in classical machine learning, statistical validation, and optimization, having developed portfolio optimization engines, regression models, and challenger frameworks across lending and fixed income use cases.My background blends engineering, quantitative modeling, and finance (MBA Gold Medalist, FRM, MS in Data Science), enabling me to bridge business problems with robust AI systems in regulated environments.Interested in applied AI, LLM systems, and intelligent automation — always open to connecting with builders and teams working on real-world AI at scale.

Experience

1. Associate Director

   UBS

   Jul 2021 — Present · Mumbai, IN

   Headed the development of ML-based champion models for Credit Risk (Pillar 2), establishing the end-to-end model lifecycle including performance monitoring, evaluation frameworks, explainability, drift detection, and periodic recalibration to meet regulatory governance standards- Architected model validity controls using unsupervised machine learning based anomaly detection (Isolation forests), dynamically identifying inconsistencies across exposures, counterparty attributes, and supporting inputs to proactively flag data quality and reporting risks- Co-developed an agentic RAG-powered Model Explainer platform, enabling credit officers to query models and documentation in natural language, reducing manual review and policy lookup effort by enabling instant, evidence-grounded answers- Embedded explainability-by-design using SHAP and LIME, surfacing feature drivers and decision rationale directly within the chatbot interface to improve transparency, auditability, and regulator defensibility- Built a “Model-as-a-Service” framework, exposing risk models as callable APIs with structured JSONschemas, enabling LLM agents and tools to perform automated reasoning, what-if simulations, and ad-hoc stress testing- Automated regulatory risk reporting through LLM agents, generating structured narratives and summaries directly from databases and model outputs, significantly reducing manual preparation time and improving consistency

Education

• Shree Navchetan School - India

  High School, Science Stream

  1997 — 2008

• Dharmsinh Desai University

  Bachelor's Degree, Electronics and communication Engineering

  2008 — 2012

• Institute of Management, Nirma University

  Master of Business Administration (M.B.A.), Finance, General

  2014 — 2016

• Northwestern University

  Master of Science - MS, Data science

Skills

• Operations Management
• Financial Services
• Business Intelligence
• Data Analysis
• Investment Banking
• Quantitative Techniques for Management
• C++
• Business Analysis
• Manufacturing
• Economics
• Microsoft Excel
• Powerpoint
• Marketing Strategy
• Training
• Management
• Valuation
• Research
• Matlab
• Project Planning
• Microsoft Office
• Financial Analysis
• Html
• Management Consulting
• Teamwork
• Team Management
• Market Research
• Project Management
• Leadership
• Managerial Accounting
• Business Strategy
• Analysis
• People Skills
• C
• Programming
• Microsoft Word
• Team Leadership