Building Cool Projects
Kunal Sachdev
Data Science | AI/ML | Backend | Full-Stack
Bachelor's in Computer Science with Minors in Statistics and Economics @ University of Waterloo.
Experience
Work Experience
Hands-on learning and real-world impact through industry placements.
2025
GenAI Engineer Intern
Edelweiss Life Insurance
May 2025 - Aug 2025 | Mumbai, India
GenAI-powered Business Insights and Visualization
- Built a serverless RAG app on AWS Bedrock (Claude 3.5 Sonnet) enabling real-time conversational access to AWS Data Lake, replacing 200+ static BI dashboards and enabling Rs. 300,000+ in estimated annual operational cost savings.
- Engineered Lambda functions for query validation, Redshift schema extraction, OpenSearch retrieval, and LLM inference; achieved 99% API uptime and reduced 8,700+ annual emails per user via automation.
- Boosted SQL generation accuracy by 28% using Cohere embeddings + OpenSearch retrieval for few-shot prompting. Integrated API Gateway + Streamlit UI, serving 1.5K+ daily active users with real-time tables/visuals.
- Fine-tuned CodeLlama-7B-Instruct using QLoRA and deployed it on NVIDIA TensorRT-LLM optimized SageMaker endpoints, accelerating inference by 1.8x and reduce memory usage by 50%.
Data Science Project - Policy Grievance and Mis-selling Prediction
- Productionized a top-5 LightGBM + CATBoost ensemble to flag grievance/mis-selling risk across 809K+ policies, surfacing 97% of grievance cases in the top decile, empowering agents to preempt complaints.
- Developed a scalable ML pipeline on AWS SageMaker, containerized with ECR, and orchestrated automated weekly predictions reports with AWS Lambda and AWS EventBridge Scheduler cron jobs, reducing prediction latency by 3x.
- Operationalized model outputs by tagging high-risk users as “Handle with Care” in internal systems and routing them to trained agents in real time, driving a 23.8% drop in 3-month rolling average of complaints.
RAGLLMNVIDIA TensorRT-LLMClaude 3.5 SonnetLlamaPrompt EngineeringAWSLambdaSageMakerBedrockHuggingFaceCohereOpenSearchStreamlitAPI GatewayPythonBashDockerLightGBMCatBoostEventBridgeECR
2024
Data Science Intern
Info Origin Inc.
May 2024 - Aug 2024 | Remote
- Deployed a custom spaCy NER model with Streamlit UI to auto-extract key resume entities, cutting HR candidate screening time by 40% and streamlining early-stage hiring workflows.
- Optimized hyperparameters for a custom 5-layer neural network using Google’s NNLM embeddings and PyTorch TensorDataset/DataLoader for efficient batching/shuffling of 2,200+ articles, achieving 96.4% accuracy in news classification.
- Conducted organization-wide technical workshops on Neural Networks, Attention Mechanisms, Transformer architectures, and LLMs.
spaCyPyTorchNamed Entity RecognitionJupyterGoogle ColabTensorFlowBERTRoBERTaScikit-LearnDoccanoStreamlitTransformersNLPStemmingLemmatizationSeabornMatplotlibBayesian Optimization
2022
Software Developer Intern
HDFC ERGO General Insurance
Apr 2022 - Jun 2022 | Mumbai, India
- Automated ingestion of insurance quote data from Excel to Oracle DB using Flask API, SQLAlchemy, Pandas, and Numpy, cutting processing time by 97.8% (1.5 hours to 2 minutes), enabling instant policy generation for underwriters.
- Implemented a 4-factor validation pipeline (file format, encrypted token, template schema, metadata check) to detect obsolete or tampered Excel templates, reducing manual entry errors and rejections by 90%.
- Designed a redundancy management system using an isactive flag for lead IDs, eliminating duplicate entries from reuploads.
FlaskOracle DatabasePandasNumPySQLALchemyAPI
Exploring Technologies
Featured Projects
Constantly learning, experimenting, and evolving with every line of code.
March•2025
Car Dealership Full-Stack Application
- Dealership Website: React frontend; Django + SQLite backend.
- Dealership and reviews service (Express.js + MongoDB + Docker), serving dealer listings/reviews via RESTful APIs.
- Reviews sentiment analyzer microservice hosted on IBM.
- CI/CD pipeline: GitHub Actions; App Deployment: Kubernetes.
December•2024
Biquadris
- C++ implementation of popular block-dropping game Tetris, enhanced with a unique 'biquadris' mode.
- Turn-based gameplay, level progression, and special block abilities.
- Object-oriented patterns (Factory Method, Observer) for modularity, extensibility, and polymorphic behavior.
- Textual and XWindows graphical interfaces.
Curious
Data Scientist
Proactive
Software Developer
Adaptable
Machine Learning Engineer
Resilient
Confident
Problem Solver
Full-Stack Developer
Curious
Data Scientist
Proactive
Software Developer
Adaptable
Machine Learning Engineer
Resilient
Confident
Problem Solver
Full-Stack Developer
About Me
A Glimpse into My World
Learn more about who I am, what I do, and what inspires me.
My Reads
Explore the books shaping my perspectives.
My Toolbox
Explore the technologies and tools I have used in my internships and personal projects.
Python
R
C
C++
Bash Script
JavaScript
HTML5
CSS3
Python
R
C
C++
Bash Script
JavaScript
HTML5
CSS3
AWS
IBM Cloud
Docker
Kubernetes
MongoDB
MySQL
SQLite
IBM Db2
Oracle
RedShift
OpenSearch
AWS
IBM Cloud
Docker
Kubernetes
MongoDB
MySQL
SQLite
IBM Db2
Oracle
RedShift
OpenSearch
Streamlit
React
Next.js
Tailwind CSS
Node.js
Express
Flask
Django
Plotly
Streamlit
React
Next.js
Tailwind CSS
Node.js
Express
Flask
Django
Plotly
Jupyter
Scikit-learn
HuggingFace
PyTorch
TensorFlow
Pandas
NumPy
SciPy
Matplotlib
Seaborn
Folium
SQLAlchemy
Jupyter
Scikit-learn
HuggingFace
PyTorch
TensorFlow
Pandas
NumPy
SciPy
Matplotlib
Seaborn
Folium
SQLAlchemy
Beyond Coding
Explore my interests and hobbies beyond the digital realm.
Fitness🏋️♂️
Poker♠️
MMA🥊
Sleep😴
Basketball🏀
Cricket🏏
Anime🎥
Badminton🏸
Food😋
Let's create something amazing together
Got a cool idea, an internship opportunity, or just want to talk about shared interests? I'm all ears - feel free to get in touch!