Listen to this article
Shaped offers a configurable platform that enhances search and recommendation systems by adapting in real-time, increasing engagement and conversion rates. Founded by Tullie Murrell and Daniel Camilleri, Shaped simplifies the complexities of data engineering, continuous training, and model observability. The platform supports seamless integration with various data sources and continuously improves through ongoing data ingestion and model retraining.
The Need for Efficient Search and Recommendation Systems
Effective search and recommendation systems are critical in today’s digital landscape. Users expect personalized experiences, yet many products fail to deliver relevant results. Common challenges include the complexity of data engineering, the constant need for model training, and issues with observability that make it difficult to monitor system performance.
Shaped aims to address these challenges by providing a configurable platform that adapts in real-time. This solution is designed to help companies enhance user engagement, conversion rates, and overall revenue through improved search and recommendation capabilities.
From a Small Project to a Game-Changer
Shaped started as a small initiative among friends, led by Daniel Camilleri and Tullie Murrell, who experienced firsthand the difficulties of building effective search and recommendation systems. During his tenure at Afterpay, Daniel encountered the intricate and time-consuming nature of creating these systems, which often resulted in less-than-adequate solutions.
Recognizing the need for a more efficient approach, Shaped was developed to simplify the process. The platform quickly grew beyond its initial scope, driven by a vision to create a user-friendly yet highly configurable product. The positive reception from the Product Hunt community highlighted the demand for such a solution, solidifying Shaped’s position as a significant player in the industry.
Key Features of Shaped: Making the Complex Simple
Shaped offers a system that adapts in real-time, increasing engagement and conversion rates. Its configurable nature allows users to tailor the system to their specific needs, ensuring relevant results.
The platform is designed for all levels of expertise, from novice developers to experienced machine learning engineers. Shaped’s intuitive interface simplifies the process of building and managing recommendation systems.
Shaped provides a variety of models that can be fine-tuned for optimal performance. The system ingests data and re-ranks it using behavioral signals, ensuring that recommendations remain relevant.
Visualizing, evaluating, and interpreting data is made easier with Shaped’s explainable results feature. The platform also boasts enterprise-grade security, with GDPR and SOC2 compliance, ensuring data is handled securely.
Recommended: NewSpring Exceeds Fundraising Goals With $390 Million Mezzanine Fund
Seamless Integration: Quick and Easy Setup with Shaped
Shaped connects directly to existing data sources, allowing for rapid integration. The process involves minimal setup, making it possible to connect data, train the first model, and fully integrate within days.
Shaped supports a variety of data sources, including Segment, Amplitude, BigQuery, PostgreSQL, MySQL, Snowflake, and more. Users, items, and events data are ingested to train and build the recommendation system.
The training process involves cross-validation on a held-out test set to select the best models. These models are then A/B tested and evaluated to ensure the highest uplift before deployment. Once deployed, the system continues to improve through continuous data ingestion and model retraining.
Daniel Camilleri’s Vision
Daniel Camilleri, Co-Founder of Shaped, shared his experiences and vision for the platform. He highlighted the complexities faced during his time at Afterpay, which inspired the creation of Shaped. The goal was to build a product that simplifies workflows for data engineering, continuous training, and model observability.
Shaped’s commitment is to be a partner in the AI journey of companies, offering a product that is not only user-friendly but also highly configurable. This dedication to improvement and innovation continues to drive Shaped forward.
The Future of Search and Recommendation Systems with Shaped
Shaped continues to innovate by staying ahead of industry trends and incorporating user feedback. The platform is committed to providing continuous enhancements, making it easier for companies to achieve relevant results quickly.
Upcoming features and improvements aim to further simplify workflows and support real-time discovery use-cases. Shaped’s partnership approach ensures that it remains a valuable tool for companies looking to build world-class search and recommendation systems.
Please email us your feedback and news tips at hello(at)techcompanynews.com
