Manchester, UK
Available for opportunities
Engineering intelligence
into production.
I'm a Senior AI Engineer with a Master's in Artificial Intelligence from the University of London. With six years of experience, I build and ship LLM-powered production systems across fintech, e-commerce, and open source.
At Verofy, I own twelve production microservices covering RAG chatbots, fraud detection, multi-provider LLM orchestration, document intelligence, and merchant risk analysis. I maintain four PyPI packages with 26,000+ downloads and publish models on Hugging Face.
Where I've worked.
Senior AI Engineer
Own twelve production microservices spanning RAG, agentic AI, fraud detection, LLM orchestration, and document intelligence for a fintech compliance platform serving 2,000+ merchants.
RAG Chatbot with Agentic Retrieval
Production RAG chatbot using Agno framework with Qdrant vector store. Handles 500+ daily queries with agentic tool-use routing. Reduced support ticket volume by 40%.
Fraud Detection & Risk Engine
CatBoost and XGBoost ensemble with SHAP explainability. Processes 10K+ transactions daily with sub-200ms inference. 94% precision at 91% recall.
Multi-Provider LLM Orchestration
Routes between OpenAI, Claude, and Grok based on task type, latency, and cost. Structured output and function calling throughout. Cut LLM spend by 35%.
Adverse Media Screening
Automated adverse media checks for fintech onboarding using LangChain. Fine-tuned RoBERTa for sentiment analysis with 90% accuracy. Halved manual processing time.
Bank Statement Extraction
Document intelligence pipeline using Claude for structured data extraction from bank statements. Handles 15+ bank formats with 96% field-level accuracy.
Merchant Monitoring Dashboard
Real-time merchant risk monitoring with Polars. Tracks transaction patterns, chargeback rates, and compliance flags across 2,000+ merchants.
Document Insights AI
Multi-modal document analysis extracting structured insights from compliance documents, contracts, and KYC materials using vision LLMs.
Documents Validation Service
Automated validation pipeline for KYC/AML documents. Cross-references extracted data against regulatory databases for compliance verification.
Merchant Assistance AI
Conversational AI assistant for merchant onboarding and support queries. Context-aware responses using RAG over merchant-specific documentation.
Merchant Pricing Analysis
ML-powered pricing intelligence analysing competitor rates, transaction volumes, and risk profiles to recommend optimal merchant pricing tiers.
Filing & Account Extraction
Automated extraction of financial data from Companies House filings and account documents. Structured output for downstream risk scoring.
Review Scraper & Sentiment
Web scraping pipeline collecting merchant reviews from Trustpilot and Google. NLP-based sentiment analysis for reputation risk scoring.
Data Scientist
Enhanced food delivery logistics and user engagement using machine learning and geospatial analysis.
ML-Powered Delivery Optimization
Improved app performance by 15% and reduced delivery times by 10%. Applied K-means clustering and geospatial analysis for route optimization.
Personalized Notification System
Analyzed user behavior using Firebase Analytics and BigQuery. Increased engagement by 20% and retention by 30% with targeted push notifications.
Data Scientist
Developed advanced ML solutions for industrial anomaly detection and financial document analysis.
Industrial Anomaly Detection
Enhanced vibrational data accuracy by 10% using LSTM networks and autoencoders. Decreased false positives by 15%. Real-time streaming via Apache Kafka.
Financial Document Analysis
Increased data extraction precision by 20% over standard OCR. Implemented TableNet and Faster R-CNN for table and form recognition.
Data Analyst Intern
Streamlined data acquisition and marketing analytics frameworks.
Automated Data Acquisition
Implemented web scraping to extract data from Google Play Console via BigQuery. Boosted marketing analytics productivity.
Content Performance Analytics
Built dashboards tracking video content engagement metrics across regions. Identified top-performing content categories, improving content curation strategy.
Data Science Intern
Implemented anomaly detection in pipeline monitoring systems.
Pipeline Anomaly Detection
Designed neural network and autoencoder models achieving 90% accuracy. Deployed Apache Kafka for real-time streaming.
Sensor Data Preprocessing
Built ETL pipeline for cleaning and normalizing IoT sensor streams. Reduced noise in vibrational data, improving downstream model accuracy by 8%.
Things I've built.
Open source tools, production apps, and side projects.
RetroCam
Browser-based vintage photo editor with retro film filters, light leaks, and grain effects.
LSE Trading Agent
Autonomous trading agent that screens FTSE 350 stocks using technical analysis, FinBERT sentiment, and LLM reasoning.
Highlightify
VS Code extension to highlight and manage important code sections with persistent, customizable colors.
What I work with.
LLMs & AI Engineering
Programming & Frameworks
Machine Learning & NLP
Cloud & Vector Infrastructure
Technical articles.
5 ML Techniques to Boost Your Model Accuracy Without Adding More Data
Discover five powerful techniques to improve machine learning model performance without collecting additional data.
Building Your Personal Portfolio Chatbot with LLMs
Unlock the power of Large Language Models to create an intelligent chatbot that showcases your professional profile.
Emotion Classification on Text using Fine-Tuned DeBERTa
Learn how to perform Emotion Classification on Text using a fine-tuned DeBERTa Model.
Sentiment Analysis on Reviews Using Fine-Tuned RoBERTa
Learn how to Perform Sentiment Analysis on Reviews Using a Fine-Tuned RoBERTa Model.
Academic background.
Master of Science in Artificial Intelligence
Royal Holloway, University of London
First Class Distinction
Certifications
Let's connect.
I'm always open to discussing AI engineering, collaboration opportunities, or interesting projects.
aglawe.ankit@gmail.com