Manav
Kheni
Selected Works
The Forge Arsenal
Gestura
Real-time ASL Translator classifying 29 signs via webcam.
Smart Customer Service Router
Automated support triage with GPT-4o-mini sentiment analysis.
DiceMail
AI-powered email platform with ultra-low latency search.
Protocol: OVERRIDE
Isometric twin-stick action game with NavMesh enemy AI.
Background
Forged in Code.
I specialize in architecting scalable AI/ML solutions and high-performance backend systems. My approach blends rigorous research with pragmatic engineering—building tools that are not only computationally sound but intuitively designed for production environments.
Whether it's optimizing recurrent neural networks for real-time edge processing or engineering automated orchestration pipelines that dramatically reduce enterprise overhead, I thrive on bridging the gap between theoretical AI models and real-world impact.
Capabilities
Technical Arsenal
Languages
AI & ML
Backend & Frameworks
Data
Databases
Cloud & DevOps
Tools
Journey
Experience
Co-Founder & Software Engineer @ DiceMail
Jan 2025 – Present
- Built an AI-powered email platform end-to-end, architecting a Node.js backend with OpenAI embeddings delivering AI-assisted search at sub-50ms latency
Software Development Intern @ Cecil Labs LLC
May – Jul 2025
- Built a multi-module PyQt5 desktop application suite with integrated AI tools and an Azure SQL-backed time-tracking platform, improving workflow efficiency by 35% and reducing manual tracking by 90%
Web Developer Intern @ Akruz Creative
Jan – May 2024
- Built and optimized 3+ Django-based web platforms, improving e-commerce functionality for 100+ active users and achieving a 17% improvement in user experience metrics
Publications
Research
Optimizing Enterprise Support Workflows: A Hybrid Orchestration Pipeline For Automated Triage And Multi-Modal Data Acquisition
Proposed a hybrid AI orchestration pipeline combining automated ticket triage and multi-modal data acquisition to reduce enterprise support overhead and improve resolution latency.
Bridging the Privacy Gap: A Localized Edge AI Pipeline for Regulatory Compliance
Designed a localized edge AI pipeline for on-device sensitive data processing, enabling GDPR/HIPAA compliance without transmitting data to external cloud services.
Connect
Ignite a Conversation
Looking to collaborate on something ambitious? Drop a message below.