Anti-money Laundering Software (AMLS)
Compiled a product PRD from scratch, defined the roadmap and sprints, and redesigned the web interface.
AI/ML/big data (fintech): AMLS ensures maximum coverage and high alerts yield.
Transaction monitoring: AMLS provides secondary scoring of the transaction monitoring process. The approach uses our proprietary semi-supervised technique which is based on a combination of multi-dimensional unsupervised techniques, network analysis, and supervised learning to detect complex money laundering structuring mechanisms.
Screening: AMLS provides secondary scoring of the screening process. The approach uses our proprietary multilayered supervised techniques that combine approaches in improved matching techniques (handles typos, spelling errors, titles, prefix/suffix, etc.) and detailed analysis of secondary information obtained either from internal sources or externally available sources.
MAESTRO: A Gauging (Metrology)/Robot Guidance Suite for the Automotive Industry
Led product design of a machine vision-based suite of products for assembly line quality assurance and robot guidance.
AI/ML/computer vision/big data (automotive): This is a complete machine vision/image processing/3D vision-based metrology product, based on cloud-of-point processing.
In addition, the solution includes a contemporary self-adaptive robot thermal compensation/absolute accuracy product and a standalone 3D vision-based metrology library (grayscale, stereo, laser projection, fringe projection).
The system was installed, among others, at Fiat (Pernambuco, Brazil) and Jaguar Land Rover (UK).
Build Verification System
Managed the design and the implementation of G-Verify: Build Verification and Validation.
AI/ML/computer vision/ big data (automotive):G-Verify: Build Verification and Validation
Bumper-to-bumper inspection and validation.
Thumbs-up for production quality based on quality assurance big data analysis, Ever-increasing expectations for vehicle customization has resulted in a myriad of complex options packages and specialized features. G-Verify solutions provide objective, reliable confirmation that all of your customer’s expectations have been successfully met.
G-Verify can be configured to work in any area of an industrial production facility:
• Bodyshop, paint shop, final assembly, moving line, stop-station
• Utilize various sensor resolutions, standoffs, illumination styles, color or gray-scale, NIR, LED, fluorescent, algorithm types, and much more to personalize a robust and reliable solution. Highly flexible, the G-Verify solution can be easily integrated into your existing production lines.
G-Verify systems offer economical and effective solutions for frame manufacturers—automatically check hundreds of holes and slots in a matter of seconds.
CAFlow | A SaaS Product That Automates the Static Code Analysis and the OSS Compliance Analysis
Designed and managed the development of a SaaS product that automates the due diligence software quality analysis.
Security/ML/big data (M&A): During the due diligence of a software product or company acquisition, the technology stack and software quality and license compliance are analyzed and reported. Additionally, maintenance, risk mitigation, and refactoring costs are automatically calculated based on a build-in big data analysis platform.
This is a SaaS product deployed on AWS that offers technology stack analysis, third-party and OSS licenses, compliance assessment, building process analysis, and a plan for the product's virtualization (migration to AWS Cloud).
G-Guide: A Robot Guidance Suite for Automotive Industry Assembly Lines
Led product design and development—G-Guide has been installed in more than 50 assembly plants (BMW, Ford, Tesla, Fiat).
G-Guide systems provide unique, innovative solutions for every automated application, from basic de-racking operations to iterative roof panel insertions to complex phased door and closure assemblies.
Return on investment and system value:
• Optimize overall assembly quality and fit
• Employ leading-edge assembly techniques
• Reduce the number of vehicles requiring rework
• Automate hazardous tasks
• Reduce the manpower dedicated to assembly operations
• Reduce the number of materials dispensed
• Eliminate precision mechanical fixturing costs
• Accommodate part and process variation
$200+ million in revenue, more than 300 installations (Ford, Daimler, Audi, Daimler, Nissan, BMW, Tesla, Hyundai, Fiat, JLR, etc.)
Shared Anti-money Laundering Typologies Framework
Defined a product roadmap and features, designed the web interface, and planned agile sprints and progress dashboards.
AI/ML/big data (fintech): Tookitakis' Shared Typologie Framework is an innovative machine learning/AI approach anti-money laundering technology. A universal cloud-based repository of anti-money laundering patterns (typologies) is automatically detected or manually compiled. It provides the ability to analyze patterns of AML criminality detected by the third party without violating PII regulations, using a data model-agnostic glass box approach. The framework mitigates the risk implied by new ML techniques.
Tech stack: AWS, Scala, Python, React
C2M -A Fully Automated Source Code Assessment Suite
Design and development of the MVP version.
Development team set up and management (100% remotely).
This is an end-to-end source code evaluation suite supports (tech-stack analysis, code risks discovery, automated code reviews, security review report, license compliance/dependency analysis/business risks report).