YANA
SCROLL
↓
Tap to See More!
SCROLL

INNOVATING

Hey! I'm

YANA

JAKHWAL

About Me

What I'm Currently Working On

Where I've Been & What I've Done

Software Engineering Intern 2025

Super.com

Built systems that power Super.com’s SEM (Search-Engine Marketing) automation, improving reliability, workflow speed, and revenue impact alongside the staff engineer.

  • Built a full ad-copy template system (backend + Redis + React) used across Google & Microsoft Ads campaigns.
    • Created smart text-shortening and URL-cleaning logic.
    • Added validation, preview features, and platform support.
  • Developed async ad-group automation workflows that replaced a slow, manual process saving ~2 hours daily.
    • Enabled growth managers to launch campaigns faster.
    • Projected to unlock ~$100K per quarter in efficiency gains.
  • Improved DSA (Dynamic Search Ads) feed syncing + alerting for Google and Bing.
    • Built accurate sync stats and surfaced them via Slack.
    • Helped teams catch failures 10× faster and protect revenue-critical feeds.
  • Contributed to Web-to-App deeplinking experiments projected to add ~$316K/year in net revenue.
  • Raised test coverage to 90%+ across SEM services to keep pipelines stable and regression-free.

Developer Intern 2025

Royal Bank of Canada

Engineered data pipelines + dashboards to streamline reporting and surface critical insights for executive visibility and decision-making.

  • Developed automated data pipelines using Python, PowerShell, MSSQL, and internal APIs to extract, transform, and deliver critical security metrics.
  • Designed scalable workflows that updated KPIs and executive dashboards in real-time, improving visibility across cryptographic infrastructure.
  • Reduced manual reporting overhead by 90% and laid foundational data pipelines for future ML-driven anomaly detection.

Machine Learning Engineer 2025

WAT.ai

Built and optimized Physics-Informed Neural Networks (PINNs) to simulate cardiac blood flow, integrating physical constraints and outperforming traditional CFD solvers in speed and scalability.

  • Designed and trained Physics-Informed Neural Networks (PINNs) to model cardiac blood flow by solving complex fluid dynamics PDEs.
  • Engineered custom loss functions and boundary condition enforcement to integrate physical constraints into neural architectures.
  • Benchmarked model performance against traditional Computational Fluid Dynamics (CFD) solvers, demonstrating competitive accuracy with significantly improved inference speed and scalability.

Director of Marketing 2025

WAT.ai

Led reading groups on cutting-edge AI/ML research and created educational content to drive community engagement and outreach.

  • Planned and ran reading groups covering cutting-edge AI/ML papers and research.
  • Created educational content including slides, videos, and interactive notebooks.
  • Drove community engagement and brand visibility through educational outreach.

Developer Intern 2024

Royal Bank of Canada

Built an ESG news summarization pipeline using RAG, HuggingFace, and LangChain, ranking 1000+ articles via vector embeddings, improving relevance by 35%, and automating reporting for stakeholders including VPs.

  • Developed an ESG news summarization pipeline using RAG with HuggingFace, LangChain, and BART.
  • Ranked 1000+ articles via vector embeddings; improved relevance with prompt tuning (+35% accuracy).
  • Automated reporting and data updates using Python, SQL stored procedures, and VBA.
  • Delivered results to 10+ stakeholders, including VPs, showcasing measurable operational impact.

Projects

ScholarFlip

A mobile application that enables students to browse, match with, and apply to scholarships through an intuitive swipe-based interface inspired by Tinder.

REPOSITORY
ScholarFlip Preview

ColdBrew

An interactive dashboard that uncovers trends and predictions for Canadian tech startups and investors through data-driven insights and visual storytelling.

REPOSITORY
ColdBrew Preview

RAG For ESG

Used NewsAPI + HuggingFace RAG pipeline to summarize ESG-related articles for investment analysis at RBC.

UNAVAILABLE
RAG for ESG Preview

Clara

A mobile application designed to assist individuals with dementia and their caregivers. Provides essential tools to enhance daily life, improve communication, and support memory retention.

REPOSITORY

Predictive Model for Supply Chain Stability

Applied MICE, HDBSCAN, Elastic Net, and Ridge Regression to analyze and model the relationship between cost of living, economic development, and supply chain resilience.

FOLDER
Predictive Supply Chain Preview

Convolutional Neural Networks Paper

Implemented and compared CNN models using TensorFlow/Keras and PyTorch on the CIFAR-10 dataset to explore deep learning architectures.

REPOSITORY
CNN Preview

Connect With Me

Email GitHub LinkedIn Substack Substack