Building intelligent systems where business, data, and AI converge
I bridge audit and finance domain expertise with deep technical skills in AI, data engineering, and backend systems—delivering production-grade solutions that drive strategic impact.
About Me
I'm a technical professional at the intersection of finance, audit, data engineering, and AI. My career began in technology audit at BlackRock, where I tested IT controls and automated testing workflows. This foundation in audit gave me a deep understanding of risk, compliance, and how enterprises manage critical systems.
At Nomura, I expanded into finance audit and analytics, working with massive datasets ($466B+ in journal entries) to identify fraud patterns and operational risks. I built dashboards that gave leadership real-time visibility into audit operations and helped drive remediation strategies.
Most recently at BDIPlus, I transitioned into pure backend and data engineering, building production-grade systems for LLM integration, multi-database orchestration, and scalable data pipelines. I architected contact extraction systems using Spark + Databricks, designed Redis caching layers, and deployed AI-powered solutions across Azure and cloud infrastructure.
Throughout this journey, I've also built production systems independently—most notably Vantage AuditOS, a zero-trust AI audit platform that automates control testing, evidence management, and SQL-driven analysis. I'm now seeking senior technical roles in Business Intelligence, Audit Analytics, or Data Engineering leadership where I can leverage both my domain expertise and engineering depth to drive strategic impact.
Quick Facts
- 5+ years across major financial and technology firms (BlackRock, Nomura, BDIPlus)
- Deep expertise in audit/finance domain + strong technical foundation in AI, data engineering, and backend systems
- Built production-ready AI audit platform (Vantage) from architecture to deployment
- Proven ability to bridge business stakeholders and technical teams to deliver measurable results
Education
MS Business Intelligence & Analytics
Stevens Institute of Technology
BE Electronics & Telecommunication
University of Mumbai
Interests
GitHub Stats
View my profile at @sahajshukla
Impact & Results
Measurable outcomes from production systems and enterprise projects
Financial Activity Analyzed
Fraud detection across journal entries
Concurrent Audits Tracked
Real-time monitoring dashboard
IT Controls Tested
SOX/ITGC compliance across 9 engagements
Latency Reduction
Redis caching optimization
Extraction Success Rate
Spark + Azure OpenAI pipeline
Manual Effort Reduced
Python automation workflows
All metrics represent real production outcomes from BlackRock, Nomura, BDIPlus, and independent projects
Trusted by Leading Organizations
5+ years across major financial services and technology firms
BDIPlus
Software Engineer II
Nomura
Associate - Finance Audit & Data Analytics
BlackRock
Analyst - Technology Audit
EliteFit AI
Backend Developer
Technical Consulting & Research
Business Intelligence Analyst Intern
Experience
My journey across major financial and technology firms, building audit automation and data analytics solutions.
Software Engineer II
BDIPlus
Jan 2025 - PresentNew York, NY
Building production-grade backend systems for data engineering, LLM integration, and enterprise data orchestration
- Architected automated contact extraction pipeline using Spark + Databricks with Azure OpenAI integration, achieving 63% extraction rate across warehouse APIs and Blob Storage
- Built proof-of-concept database backend for ChatCDP platform, deployed for Morgan Stanley client demos
- Engineered LLM prompt orchestration system using Azure OpenAI + Gemma-3 8B across multi-environment deployments
- Refactored monolithic Python/Spark workflows into scalable C# dependency injection-based architecture, improving maintainability and testability
- Designed and implemented Redis JSON caching layer for large data objects, reducing query latency by 40%
- Integrated multi-database systems (Snowflake, PostgreSQL, SQL Server) with unified data access layer
Associate - Finance Audit & Data Analytics
Nomura
Apr 2024 - Jan 2025New York, NY
Led analytics for finance audits, building fraud detection systems and operational risk dashboards
- Conducted fraud detection analytics across 160,000+ journal entries representing $466B in financial activity, identifying high-risk patterns for audit follow-up
- Designed and deployed 5-year operational risk dashboard for Global Finance Audit team, providing leadership with real-time visibility into audit metrics
- Supported 3 major Finance Audits: Financial Reporting, Collateral Management, and Dodd-Frank Compliance
- Performed root cause analysis on audit findings, identifying top remediation priorities for executive leadership
- Built automated exception testing workflows using Python and SQL, reducing manual review time by 35%
Analyst - Technology Audit
BlackRock
Aug 2021 - Jan 2024New York, NY
Evaluated IT controls and built automation tools for SOX/ITGC compliance testing
- Designed and deployed Python-based automation for IT control testing, reducing testing time by ~6 hours per control and improving audit efficiency
- Built real-time monitoring dashboard tracking ~200 concurrent audits, providing visibility into audit progress and resource allocation
- Tested design and operating effectiveness of 65 IT controls across 9 audit engagements covering applications, infrastructure, and cybersecurity
- Co-authored 4 formal audit reports for senior management and external auditors, documenting control deficiencies and remediation plans
- Identified and remediated 7 critical control issues, working with IT teams to implement fixes and validate effectiveness
- Developed business overview analytics using AUM and risk data to contextualize audit findings for stakeholders
Backend Developer
EliteFit
Aug 2020 - Mar 2021Singapore (Remote)
Built ML-powered fitness form correction system and backend API infrastructure
- Developed REST API endpoints connecting React frontend to Python ML backend for real-time pose analysis
- Built pose-scoring ML system for fitness form correction achieving 93% accuracy vs. baseline models, helping users improve workout technique
- Collaborated with cross-country teams (Singapore, US) for testing, deployment, and feature iteration
- Deployed ML model assets and user data to AWS S3 and DynamoDB for scalable cloud infrastructure
Business Intelligence Analyst Intern
Technical Consulting & Research Inc.
Mar 2020 - Aug 2020Newark, NJ
Built real-time analytics pipeline and automated reporting infrastructure
- Designed and deployed analytics pipeline using REST APIs + Kafka for real-time data ingestion and processing
- Built text classification model with 88% accuracy for sentiment analysis and trend detection from customer feedback
- Automated reporting workflows using AWS Lambda + Google Data Studio, reducing manual effort and improving net profit visibility by 57%
- Created executive dashboards providing business leaders with actionable insights into operational metrics
Certifications
Professional certifications and accreditations
Skills & Expertise
A comprehensive skill set spanning audit, analytics, engineering, and cloud technologies.
Always learning and expanding my skill set. Currently exploring LLM fine-tuning, vector databases, and advanced RAG architectures.
Projects
Building production-grade systems that solve real-world audit and data challenges.
Vantage AuditOS
AI-Powered, Zero-Trust Audit Automation Platform
A production-ready audit management platform combining AI-driven SQL generation, secure evidence management, and real-time control testing workflows.
Vantage AuditOS is a full-stack audit automation platform I architected and built to solve real-world challenges in technology and finance audit. The platform is production-ready and demonstrates enterprise-grade engineering practices.
The system uses a FastAPI microservices backend, Next.js frontend, and a self-hosted LLM (OpenSQL-7B) to enable auditors to perform complex analyses through natural language queries. This eliminates the need for auditors to write SQL manually while maintaining full transparency and auditability.
Security is built into the foundation: per-audit AES-256 encryption with isolated key management, Postgres-based relational data model for audits/controls/workpapers/findings, and complete workflow automation for evidence ingestion, control testing, workpaper generation, and multi-level approval routing.
The platform is designed with zero-trust principles—every API call is authenticated, all data at rest is encrypted, and comprehensive audit trails are maintained for compliance. Vantage represents the convergence of my audit domain expertise and technical engineering skills.
Tech Stack
Key Features
- Natural language to SQL generation for audit queries using self-hosted LLM
- Per-audit encryption with isolated key management for multi-tenant security
- Evidence upload pipeline with automated validation and secure storage
- Automated control testing engine with exception detection and risk scoring
- Workpaper generation system with approval workflows and version control
- Real-time collaboration features with comprehensive audit trail logging
System Architecture
Hover over components to learn more about the architecture
Other Projects
Neural Style Transfer (PyTorch)
Deep learning image transformation using Gram matrices
Implementation of neural style transfer using PyTorch, applying artistic styles to photographs through learned feature representations.
- Learned and implemented core deep learning concepts (feature extraction, optimization)
- Successfully generated Van Gogh-styled portrait as personalized gift
- Demonstrates understanding of CNN architectures and transfer learning
FIFA Player Recommendation System
ML-powered team building with budget constraints
Built recommendation system using PCA and kNN clustering to help fantasy football managers build optimal teams within budget constraints.
- Demonstrates applied ML for optimization problems
- Shows understanding of dimensionality reduction and clustering techniques
- Real-world application of constrained optimization
Baby Cry Predictor
Audio classification using sound wave analysis
Multi-class classifier that analyzes baby cry audio to predict cause (hunger, discomfort, tiredness, etc.) using sound wave features.
- Shows expertise in audio signal processing and feature extraction
- Demonstrates ability to work with non-traditional data types
Weather App - AWS EKS Kubernetes Deployment
Cloud-native full-stack application on AWS
Three-tier React/Node.js/MongoDB application deployed on AWS EKS with Kubernetes orchestration, demonstrating cloud-native architecture skills.
- Demonstrates end-to-end cloud deployment skills
- Shows understanding of containerization and orchestration
- Production-grade infrastructure management
COVID-19 Statistics Dashboard
Real-time pandemic tracking dashboard
Interactive jQuery-based dashboard providing worldometer-style COVID-19 statistics with real-time updates and visualizations.
- Shipped quickly during crisis to provide public value
- Demonstrates ability to build practical tools under time pressure
Racing Lyrics - 4.7M Songs Visualization
Animated language comparison across millions of songs
Creative data visualization comparing languages across 4.7 million songs using animated racing-line charts.
- Shows ability to work with large datasets creatively
- Demonstrates data storytelling and visualization skills
Sudoku Solver (Backtracking Algorithm)
Classic algorithms implementation
Recursive backtracking algorithm implementation for solving Sudoku puzzles, demonstrating algorithmic thinking and optimization.
- Shows strong foundation in classical computer science algorithms
- Demonstrates problem-solving and optimization skills
Writing
Exploring the mathematical foundations of data science, statistics, and information theory.
Open to New Opportunities
I'm actively seeking senior technical roles where business intelligence, domain expertise, and technical depth converge. I bring 5+ years of experience across audit, data, and engineering—with a proven track record of building production systems and driving measurable business impact.
Target Roles
What I Bring to Your Team
- 5+ years across major financial services and technology firms (BlackRock, Nomura, BDIPlus) with deep audit/finance domain expertise
- Strong technical foundation in AI/ML, data engineering, and backend systems—proven ability to architect and deploy production-grade platforms from scratch
- Bridge business stakeholders and technical teams effectively—translate complex requirements into scalable solutions that drive strategic impact
- Track record of automation and efficiency gains: built systems that save hours per control test, reduce manual effort by 35-60%, and improve operational visibility
Get In Touch
I'm always open to discussing new opportunities, collaborations, or just chatting about data, audit, and AI. Feel free to reach out!
Direct Contact
I typically respond within 24 hours. For urgent opportunities, feel free to call me directly.