Thank you for taking the time to share your thoughts on our interview process. We would like to address your concerns and provide context for why we approach our hiring process the way we do. Why We Use Technical Assessments At Rhombus AI, we believe that technical assessments are the best way to evaluate a candidate's abilities in a real-world context. By focusing on a real-world application, we ensure that each candidate is assessed on their ability to deliver practical, impactful solutions. This is a significant departure from many companies that simply screen applications based on CVs or conduct surface-level coding challenges. In fact, many candidates have shared with us that they love our assessment process because it allows them to demonstrate their coding and problem-solving skills directly. It levels the playing field and provides a unique opportunity to highlight their technical capabilities, creativity, and approach to solving complex problems. Addressing Your Concerns Assessment Scope We understand that creating a full-stack application using Django and React within two weeks is demanding. However, this challenge is designed to simulate real-world scenarios that reflect the complexity of our projects. Some outstanding candidates successfully navigated this challenge while incorporating additional features, such as: Setting up a CI/CD pipeline for automated deployments. Adding distributed tracing systems (e.g., Jaeger) to debug complex workflows. Integrating advanced logging and monitoring solutions using Prometheus, Grafana, and ELK. These enhancements go beyond the base requirements and reflect an understanding of modern development practices. Feedback on Data-Type Inference Candidates who excelled used innovative methods such as: Clustering Techniques: One candidate implemented a DBSCAN-based algorithm to better infer categorical data by analyzing distribution density in multi-dimensional space. Categorical Identification Using Vectors: Another candidate analyzed clustering behavior in a 2D vector space to improve detection accuracy. Handling Big Datasets Scalability is essential in our work. Outstanding candidates demonstrated strategies such as: Reading files in chunks to avoid memory overload. Asynchronous processing (e.g., Celery) to handle files exceeding 3 GB without performance issues. Front-End Observations Outstanding candidates demonstrated a balanced skillset incorporating advanced features like React Context, zustand for state management, and Server-Sent Events (SSE) for real-time updates. Cloud Computing and Architecture We also evaluate candidates on their ability to design cloud-ready applications. One standout submission showcased: A state-machine-driven backend using libraries like Transitions to manage process states effectively. Integration of AWS S3 for presigned URLs to offload file upload responsibilities. Regarding Communication Delays We acknowledge and apologize for the delay in communicating our decision. With over 350 applications, we aimed to provide detailed and personalized feedback rather than a generic response. This required additional time, but we believe every candidate deserves a thoughtful evaluation of their work. Closing Thoughts Through this rigorous process, we were fortunate to hire four extremely dedicated, hardworking, and motivated software engineers. Their contributions have enabled us to scale exponentially and bring our web app to market at an accelerated pace. This success reinforces the value of our approach to hiring and underscores our commitment to finding and supporting exceptional talent.