Idhant Das

Architected a high-performance, real-time content moderation pipeline for multimodal (text + image) user inputs, integrating threshold-based classification, asynchronous validation queues, and per-request filtering logic, enforcing strict safety policies while preserving sub-second latency and seamless user experience under growing production load Engineered comprehensive application observability using Firebase Crashlytics augmented with custom structured logging, distributed error tracing, and alert thresholds, slashing mean-time-to-resolution (MTTR) by 35% and accelerating root-cause debugging across live iOS production builds Managed iOS beta release workflows through TestFlight for 20+ testers, coordinating feedback cycles and improving deployment iteration speed by 40% prior to App Store releases

Software Engineer | Full Stack at OurFreedom.ai (2026-01 – Present)

• Architected a high-performance, real-time content moderation pipeline for multimodal (text + image) user inputs, integrating threshold-based classification, asynchronous validation queues, and per-request filtering logic, enforcing strict safety policies while preserving sub-second latency and seamless user experience under growing production load
• Engineered comprehensive application observability using Firebase Crashlytics augmented with custom structured logging, distributed error tracing, and alert thresholds, slashing mean-time-to-resolution (MTTR) by 35% and accelerating root-cause debugging across live iOS production builds
• Managed iOS beta release workflows through TestFlight for 20+ testers, coordinating feedback cycles and improving deployment iteration speed by 40% prior to App Store releases

SWE Intern | Backend Developer at DCUBE LLC (2023-07 – 2024-08)

• Led a 6-person engineering team to develop an AI paralegal assistant that converts case descriptions and evidence into vectors to retrieve and cross-reference relevant legal precedents, cutting research and drafting time to a fraction of the traditional labor
• Built ML-based RAG (Retrieval-Augmented Generation) pipelines for legal text with LangChain, Redis/RediSearch, and Azure AI, enabling semantic search and structured citation retrieval for users via chatbot interface
• Reduced query latency and improved precision by 45-60% through deployment of a Redis vector database for scalable document embeddings and cosine-similarity retrieval
• Designed API endpoints and backend systems for document encoding, query expansion, and large-scale data storage in Python

SWE Intern at PatientXpress (2024-06 – 2025-09)

• Designed and deployed scalable REST APIs in Node.js and PostgreSQL to support scheduling and billing operations, improving system consistency and reducing billing discrepancies by 30% across multiple clinic workflows
• Optimized database performance by analyzing query execution plans and implementing indexing strategies, reducing application latency by 61% and improving responsiveness for thousands of active users
• Developed automated regression testing pipelines and structured debugging workflows, increasing system stability and contributing to an 40% improvement in customer satisfaction metrics

Machine Learning Researcher at Living Matter Lab (2024-10 – 2025-01)

• Engineered a high-performance 2D Active Brownian Particle simulator with 25M+ force evaluations/run, accelerated by 75% using parallelization
• Automated analysis pipelines with Python/HDF5, processing 4 GB+ datasets and revealing a 15% faster phase separation under noise
• Packaged simulation tools and metrics (MSD, RDF, velocity distributions) into a reproducible, modular research codebase used for publication

B.S. in Computer Science – University of California, San Diego (2022-09 – 2025-09)