Karamvir Singh
Verified Expert in Engineering
Data and AI Engineer and GenAI Developer
Toronto, ON, Canada
Toptal member since August 14, 2025
Karamvir is a data and AI engineer and a Generative AI (GenAI) expert with 6+ years of experience in Python, AI/ML, and cloud platforms like Google Cloud Platform and Amazon Bedrock. He builds scalable, large language model (LLM) powered solutions, from trading bots to automated content systems and RAG pipelines, leveraging cutting-edge tools like LangGraph, CrewAI, and Amazon Bedrock. Karamvir turns complex AI research into high-impact, production-ready products that drive real business value.
Portfolio
Experience
- Python 3 - 7 years
- LangGraph - 3 years
- Generative Artificial Intelligence (GenAI) - 3 years
- Google Cloud Platform (GCP) - 3 years
- Pinecone - 2 years
- FastAPI - 2 years
- OpenAI API - 2 years
- Retrieval-augmented Generation (RAG) - 2 years
Preferred Environment
Python 3, AWS Bedrock AgentCore, LangChain, LangGraph, Amazon Web Services (AWS), Retrieval-augmented Generation (RAG), Model Context Protocol (MCP), Machine Learning Operations (MLOps), Docker, Google Cloud Platform (GCP), OpenAI, AI Chatbots, Chatbots, ChatGPT, Agentic AI, Automation, n8n, Tutoring, Machine Learning, Python, Recommendation Systems, Natural Language Processing (NLP), Pinecone, Vector Search, Agentic Frameworks, Claude Code, LlamaIndex, RAG Systems
The most amazing...
...thing I've built is an LLM-powered music content marketing system that boosted artist streams and revenue more than ten times in the first month of use.
Work Experience
eCommerce Search AI Back-end Developer
r2decide Inc.
- Architected and delivered the platform’s core retrieval + ranking engine, combining Pinecone vector search, metadata filters, and LLM-driven query intent extraction/moderation to enable hyper-personalized shopping.
- Built and hardened our back-end stack end-to-end—FastAPI services, async pipelines, Redis caching, rate limits, fallbacks, and monitoring—ensuring the product stayed fast and reliable through rapid iteration.
- Designed and operationalized personalization pipelines that captured real user purchase signals and fed them into ranking, directly improving relevance.
GenAI Engineer
Save the Dev
- Leveraged LLM image analysis capabilities to develop strong price move detection for the WGD trading bot, increasing accuracy by 30% over standard algorithms like ATR.
- Packaged the LLM-based strong move detection into a Model Context Protocol (MCP) server and connected it to a multi-agent financial analysis and trading system.
- Developed a full-scale auto content posting system for music promotions based on song lyrics using LLMs for ideation and content generation and LangGraphs for workflow orchestration.
Int. Data Engineer
Loblaw Digital
- Analyzed the Ontario Loblaws drug stores' data to develop an automatic store-to-store drug moving recommendation system, reducing drug disposal costs by 30% in Loblaw-owned stores in Ontario.
- Modernized and automated project data pipelines, achieving five times speedups, reducing cycle times from over 50 hours to just 10 hours, and handling 2+ million records per cycle using Jupyter, BigQuery, and Vertex AI.
- Collected project requirements, built and maintained datasets and SQL queries, and developed and deployed the primary AI model using AWS Services.
ML Operations Engineer
Mia Platform
- Designed and implemented the end-to-end auto model deployment pipeline using Vertex AI, AWS Sagemaker, Docker, and Kubernetes.
- Served as a subject matter expert (SME) for the co-founders and other developers, providing guidance on ML, MLOps, Google Cloud, and Kubernetes..
- Assisted the company in pivoting toward GenAI consulting and built four GenAI apps using GPT API, including timesheet SRED organizer, eCommerce product classifier, transport invoice anomaly detector, and invoice product info extractor for Shopify.
AI Strategy Consultant
Lemay.ai
- Developed predictive models and data science solutions for 3+ Lemay clients using random forests, XGBoost, spaCy, HuggingFace, Bert, TensorFlow, and scikit-learn.
- Designed the SEC database's data ingestion and querying infrastructure with over 160 million rows using MySQL and AWS.
- Optimized the architecture to achieve over tenfold efficiency in data processing pipelines, e.g., from two weeks to 18 hours, using Pandarallel, NLP pipe, and Python multiprocessing.
President
GESPDC, University of Ottawa
- Presided over a 700+ graduate engineers' club working toward their professional development.
- Led a team of seven PDC executives and 10 program coordinators working toward the professional development of University of Ottawa engineering graduates.
- Built and strengthened working relationships with 15+ industry and community partners, including the City of Ottawa, Ottawa Public Health, Bank of Canada, NRC, FBSC, ANQWT, etc.
AI Engineer
City of Ottawa, PDC UO
- Developed an end-to-end computer vision application to guide waste classification following the City of Ottawa policies.
- Created a dataset of 30,000+ waste item images, from various sources like Google Images, Kaggle datasets, and student surveys.
- Built, trained, and compared multiple models, including VGG-16, ResNet50, Inception V3, and AlexNet, for image-to-bin classification with 80-90% accuracy.
- Created a Python Web API using FastAPI and deployed it on AWS using Amazon EC2 and Docker.
- Developed a Flutter app and integrated it with the model.
IT Analyst
KPMG
- Worked on four different client-facing SOC reporting engagements in teams of four to 10.
- Gathered, documented, and analyzed client process data for 150 to 200 process controls per project using Advanced Excel functions and macros.
- Proposed, initiated, and implemented automation procedures using VBA and Office macros, speeding up the internal documentation by 50 times.
Experience
City of Ottawa Waste Management AI
Oanda Tradebot with GenAI
RAG-based Chatbot for Development Shop
eCommerce Back-end Pipeline
• Architected the end-to-end retrieval and ranking stack, combining vector search (Pinecone), hybrid metadata filtering, and LLM-based query understanding to convert natural language input into structured intents and high-relevance results.
• Focused on speed and accuracy, targeting 0.7+ f1 scores and sub 500 ms responses.
• Developed scalable FastAPI services and async pipelines for search, recommendations, and user memory, with integrated caching (Redis), rate limiting, retries, and fallbacks to ensure low latency and resilience during rapid product iteration.
• Designed and implemented the personalization and data-signal layer, capturing engagement events—including searches and purchases—and feeding these into downstream ranking systems to continuously improve relevance.
• Worked closely with research on retrieval evaluation and with product/design teams to refine user-facing experiences, enabling fast experimentation and reliable system performance across the launch.
Education
Master's Degree in System Science
University of Ottawa - Ottawa, Canada
Master's Exchange in Quantum Computing
University of Waterloo - Waterloo, Canada
Bachelor's Degree in Computer Science and Technology
Guru Nanak Dev University - Amritsar, India
Certifications
Deep Learning Nanodegree
Udacity
Skills
Libraries/APIs
OpenAI API, REST APIs, Pandas, TensorFlow, PyTorch, Python Asyncio, SQLAlchemy
Tools
n8n, ChatGPT, Claude, Claude Code, Jupyter, Microsoft Excel, Amazon Simple Queue Service (SQS), AWS IAM, Amazon Simple Notification Service (SNS), Uvicorn
Languages
Python 3, Python, SQL, Dart, Java, C++, MQL5, Excel VBA
Frameworks
LangGraph, Agentic Frameworks, LlamaIndex, Starlette, Flutter
Paradigms
Model Context Protocol (MCP), Automation, Object-oriented Programming (OOP), DevOps, Agile Software Development, ETL
Platforms
Shopify, Ollama, Amazon Web Services (AWS), Docker, Google Cloud Platform (GCP), Vertex AI, Kubernetes, Amazon EC2
Storage
Databases, RDBMS, Google Cloud, Amazon S3 (AWS S3), Amazon DynamoDB, PostgreSQL
Other
LangChain, Retrieval-augmented Generation (RAG), Artificial Intelligence (AI), Generative Artificial Intelligence (GenAI), Agentic AI, Gemini API, API Integration, Machine Language, OpenAI, Workflow Automation & System Integration, Back-end, Large Language Models (LLMs), Natural Language Processing (NLP), Vector Databases, AI Agents, Food, Shopify Design, Front-end Development, Experiential Design, Command-line Interface (CLI), Architecture, Software Architecture, AI Chatbots, Chatbots, Multimodal Models, Tutoring, Pinecone, Semantic Search, Vector Search, Solution Architecture, Data Science, Large Language Model Operations (LLMOps), Agentic RAG Systems, Workflow Automation, RAG Systems, Amazon Bedrock AgentCore, Machine Learning Operations (MLOps), Machine Learning, APIs, FastAPI, Professional Development, Big Data, System Design, Deep Learning, Data Engineering, Recommendation Systems, Multimodal GenAI, Leadership, Remote Team Leadership, Mentorship & Coaching, Data Analysis, Knowledge Bases, Amazon API Gateway, Amazon App Service, ChatGPT API, ServiceNow, ElevenLabs Solutions, Anthropic, Learning Management Systems (LMS), Computer Vision, Quantum Computing, Quantum Mechanics, Trading Bots, TradingView, Event Management, Event Marketing, Mobile Apps, Quantization, SOC 2, Compliance, IT Audits, Macros, Excel Macros, Lead Generation, Sales Funnel, Groq, Voyage AI
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring