YK
Yasin Khan
Contact details — on requestProof of Work — on request
Experience
Senior Consultant
HCLTech · Sep 2025 – Present
- Designed and deployed Machine learning model, conversational AI and multi-agent automation systems integrated with analytics dashboards using GCP.
- Led a team of 9 members and delivered scalable enterprise AI solutions.
- Developed a Proof-of-Concept (POC) anomaly detection system to identify fraudulent claims and suspicious transaction patterns for a medical insurance client (POC Anomaly Detection for Financial Fraud).
- Performed exploratory data analysis and engineered features for anomaly detection.
- Built unsupervised and supervised anomaly detection models and integrated output into risk dashboards.
- Outcome: Early identification of suspicious claims, reduced false positives, improved fraud investigation efficiency.
- Developed an AI-driven solution to detect anomalies in carbon emission patterns and provide Root Cause Analysis (POC Carbon Emission Reduction Model).
- Implemented Hybrid ML + RAG-based Framework with Isolation Forest, Auto-thresholding, Random Forest Regressor.
- Outcome: Identified root causes, provided data-driven sustainability recommendations, enabled proactive emission monitoring.
- Developed enterprise chatbot using Dialogflow CX integrated with Looker dashboard data (Conversational Chatbot).
- Designed intent flows, entity extraction, and contextual conversation handling, integrated APIs for live metrics.
- Outcome: Automated dashboard insights retrieval and reduced manual reporting dependency.
- Designed and implemented Multi-Agent architecture using ADK for ticket management automation (Multi-Agent System).
- Built agents for ticket assignment, status tracking, escalation handling, resource allocation with automated logic.
- Outcome: Improved ticket resolution efficiency and optimized workforce allocation.
- Deployed chatbot and multi-agent system using Google Cloud App Engine, ensured scalability and high availability (Deployment & Infrastructure).
- Managed a team of 9 AI/Engineering professionals, overseeing architecture, development, code quality, deployment, and stakeholder coordination (Team Leadership).
Senior Data Scientist
Capgemini · Jul 2024 – Sep 2025
- Designed predictive and NLP-based machine learning solutions across aviation, pricing intelligence, and customer analytics domains.
- Led a team of 8 members and delivered scalable ML systems deployed on AWS.
- Built recommendation model based on user purchase history and browsing behavior, applying Collaborative Filtering, Content-Based Filtering, Matrix Factorization (Recommendation System Retail Customer Behavior).
- Engineered behavioral features and integrated recommendation API into retail platform.
- Outcome: Improved personalized product suggestions and increased cross-sell opportunities.
- Implemented Retrieval-Augmented Generation (RAG) for HR policy and documentation search using vector embeddings, similarity search, and a context-based retrieval pipeline (RAG Implementation HR Documentation System).
- Integrated document indexing and chunking strategy for accurate responses.
- Outcome: Reduced HR query resolution time and improved internal knowledge access.
- Developed chatbot leveraging OpenAI LLM, AWS Bedrock models, and vector database for retrieval (Enterprise Chatbot OpenAI + AWS Bedrock).
- Designed multi-step retrieval pipeline for contextual response generation and deployed scalable API-based chatbot solution.
- Outcome: Automated enterprise Q&A support and reduced manual dependency.
- Built GenAI-based summarization system for large call and chat transcript data, implementing prompt-engineered summarization workflow using LLM APIs (Text Summarization Call & Chat Transcripts).
- Automated structured summary generation for analytics and reporting.
- Outcome: Improved operational visibility and reduced manual review effort.
- Designed and implemented Agentic AI architecture using ADK for real-time automotive chatbot solution (Agentic AI Solution Real-Time Automotive Support).
- Developed multi-agent orchestration for real-time webhook data ingestion, live database queries, and context-aware response generation.
- Implemented Agent-to-Agent (A2A) communication, built multilingual conversational agent, and deployed scalable cloud-based API solution.
- Outcome: Enhanced customer experience with real-time vehicle status tracking and reduced service center workload.
- Managed and mentored a team of 4 ML/AI engineers, overseeing model design/validation, deployment planning, code reviews, and stakeholder communication (Team Leadership).
Software Delivery Engineer
Mphasis · Feb 2022 – Jul 2024
- Designed predictive and NLP-based machine learning solutions across aviation, pricing intelligence, and customer analytics domains.
- Led a team of 8 members and delivered scalable ML systems deployed on AWS.
- Built passenger no-show prediction model using Logistic Regression, Random Forest, XGBoost, Deep Neural Networks (Predictive Modeling Flight Overbooking Optimization).
- Engineered behavioral and booking pattern features, developed probability-based optimization framework, and deployed model using AWS SageMaker endpoints.
- Outcome: Increased seat utilization, improved revenue optimization, reduced denied boarding risk.
- Collected and processed Twitter data for airline/customer sentiment analysis, implementing TF-IDF + Logistic Regression, LSTM, BERT-based sentiment classification (Sentiment Analysis & Social Media Intelligence).
- Built interactive dashboard using AWS QuickSight for real-time monitoring.
- Outcome: Enabled brand sentiment tracking and early issue detection.
- Developed intelligent auto-complete suggestion model for customer feedback forms, implementing N-gram, LSTM, Transformer-based text generation (NLP - Auto-Complete Feedback System).
- Outcome: Improved user experience, increased feedback submission rate, enhanced customer engagement and data collection efficiency.
- Built regression-based pricing models using data scraped from multiple websites, applying Linear Regression, Random Forest Regressor, XGBoost (Price Prediction & Competitive Intelligence).
- Engineered time-series and competitor-based features, developed dynamic pricing insights dashboard.
- Outcome: Improved pricing accuracy and competitive positioning.
- Led a team of 8 data scientists and engineers, managing model development lifecycle, code reviews, deployment planning, and stakeholder communication (Team Leadership & Project Management).
- Ensured delivery within timeline and business KPIs, successfully delivered multiple production-grade ML solutions.
H
GIS Analyst 1
HERE Technologies · Jan 2019 – Feb 2022
- Possessed 2 years and 10 months of experience in GIS Analyst & Machine Learning Engineer with expertise in geospatial intelligence, computer vision for ADAS, route optimization, customer analytics, and scalable ML deployment on AWS.
- Delivered AI-driven transportation and location intelligence solutions that reduced operational cost and improved safety.
- Developed real-time traffic sign detection model using YOLO (v5/v8), applying CNN-based deep learning and transfer learning (Computer Vision Engineer - ADAS Traffic Sign Detection).
- Improved object detection accuracy through data augmentation and hyperparameter tuning, deployed model using AWS SageMaker endpoints.
- Outcome: Enhanced ADAS safety features with low-latency detection.
- Built intelligent route optimization engine using Dijkstra & A*, Random Forest / XGBoost for cost and fuel prediction (Geospatial ML Engineer - Route Optimization & Emission Reduction).
- Designed weighted graph models for road network analysis, integrated emission modeling, and exposed routing service via API using AWS Lambda & API Gateway.
- Outcome: Reduced fuel cost, improved logistics efficiency, supported ESG goals.
- Performed geospatial clustering using K-Means, DBSCAN, PCA for dimensionality reduction (Data Scientist - Customer Segmentation ADAS Users).
- Segmented customers based on driving behavior and ADAS usage patterns, enabled targeted marketing and feature personalization.
- Outcome: Improved customer targeting and feature adoption rate.
- Applied Apriori & FP-Growth algorithms for association rule mining (Data Scientist - Market Basket Analysis).
- Identified frequently co-purchased ADAS features, supported cross-selling and bundling strategy.
- Outcome: Increased bundled product revenue and recommendation accuracy.
- Developed Sentiment Analysis models using TF-IDF + Logistic Regression, LSTM, BERT (NLP & Deep Learning Engineer).
- Implemented NER & Topic Modeling (LDA) for geospatial document insights, automated customer feedback analysis pipeline.
- Outcome: Enabled real-time customer insight extraction.
- Performed model training & tuning using AWS SageMaker, managed real-time inference endpoints, batch processing via AWS Lambda, and model monitoring using CloudWatch (Model Deployment & MLOps).
- Utilized Docker containerization & CI/CD pipelines.
Skills 0 proven through work
Also works with
LLM Output ParsingTwitter Data AnalysisMetadata QueryingGPT-2.5 Flash Lite UsageGPT-3.5 Flash UsageTechnical TroubleshootingProblem SolvingInformation ResearchWork Planning and PrioritizationDashboard UI DesignKeyword DetectionSentiment AnalysisNLP Text AnalysisPrompt EngineeringAmazon S3AWS DeploymentClassification ModelingPredictive MaintenanceGemini AI IntegrationDocument ChunkingMultimodal Data ProcessingApplication DeploymentText EmbeddingInternet Search IntegrationEvent-Driven ArchitectureServerless ArchitectureObject Storage ManagementChatbot DevelopmentRetrieval-Augmented GenerationAI Agent DevelopmentGCP Tool IntegrationMulti-Agent System IntegrationConversational UX DesignMachine Learning ApplicationArtificial Intelligence Fundamentals
Proof of Work
Proof of Work
Yasin shares this with people who ask. You'll hear back either way.
Education
MS/M.Sc(Science), Computers
Mumbai University · 2023 — 2025
BCA, Computers
Pune University · 2013 — 2016
12th, English
Maharashtra · 2011 — 2012
10th, English
Maharashtra · 2009 — 2010
Contact details
Contact details
Yasin shares this with people who ask. You'll hear back either way.
Ask about this candidate
AI responses may contain errors. Proof status and source information are provided by Proof of Skill.
A Proof CV — one profile, kept current, with the proof attached. Back to the portal