← Back to Projects

Gusto

Graph AI

Heterogeneous graph neural network for preference-aware entity matching. GNN message-passing over multi-relational interaction graphs with hybrid ranking. Cold-start mitigation across sparse interaction regimes.

PythonGNNPyTorchReactFlaskPostgreSQLGraph NetworksRecommendation Systems

About

Heterogeneous graph neural network for preference-aware entity matching. GNN message-passing learns preference embeddings through neighborhood aggregation. Hybrid ranking combines structural scores with content-based features.

The Problem

Traditional recommendation relies on aggregate signals. Collaborative filtering suffers from cold-start and data sparsity. The interaction graph contains rich preference signals that flat models ignore.

The Approach

Heterogeneous graph construction with representation fallback. GNN message-passing learns preference embeddings. Hybrid ranking weighted dynamically based on interaction density. Content-based fallback for cold-start.

Tech Stack

  • Frontend: React, JavaScript, TailwindCSS
  • Backend: Python 3.11, Flask, PostgreSQL, Redis
  • AI/ML: PyTorch, PyTorch Geometric, Scikit-learn, Collaborative Filtering

Apply

Apply by April 30, 2026

Slack: #gusto

Apply Now →

You'll learn

  • Graph Neural Networks
  • PyTorch Geometric
  • Recommendation Systems
  • Collaborative Filtering
  • Cold-Start Problem

Roles

Rashanjot Kaur - ML Architect

Designed heterogeneous GNN architecture and cold-start mitigation strategy.

6 open role(s)

Applied AI Engineer - Open

Train and refine PyTorch Geometric models, hybrid ranking, and collaborative filtering integration.

Apply →

Backend Engineer - Open

Build graph construction pipeline, Flask APIs, PostgreSQL/Redis serving, and inference integration.

Apply →

Frontend Engineer - Open

React UI for graph exploration, rankings, and cold-start explainability.

Apply →

Data Engineer - Open

Graph construction from interaction logs, feature stores in PostgreSQL, and data quality for GNN inputs.

Apply →

AI Researcher - Open

GNN and cold-start experiments, offline evaluation methodology, and hybrid ranking studies.

Apply →

DevOps / MLOps - Open

Training and deployment pipelines, Redis caching, monitoring for inference and batch graph builds.

Apply →