ContextQA
AI Engineer
- Working on ContextQA's model infrastructure and evaluation systems.
- Building multimodality support and voice infrastructure for agents.
I build production-ready AI systems across agents, model workflows, computer vision, and full-stack products.
Experience
AI Engineer
Core AI Engineer
AI/ML Systems & B2B Consulting
Audience Growth & Sales Funnels
Few projects

AI system for iOS/Android
Real-time TCG card scanning for a collector platform with 50K+ users, tuned for motion blur, perspective, and tiny ID regions.
Computer vision demo · 90%+ accuracy
Sports CV pipeline with 90%+ accuracy for player tracking, pose, ball movement, and stable overlays from match footage.
Security model fine-tuning
Fine-tuned OpenHermes and Llama for safer, vulnerability-aware responses within a broader AI infrastructure system.
Research
Published work on micro-expression recognition using Swin-T on CASME II, comparing spatial and temporal heads, with gains from temporal cues and RAF-DB pretraining.
Writing
Open source
SGLang Project - SpecForge
Fixed loss-mask alignment during target preprocessing, restoring missing target supervision and removing stale boundary gradients. Added regression coverage for mask edge cases.
BrainDrive - ModelMatch
Built and maintained an LLM-as-judge evaluation harness for comparing open-source models across finance, health, summarization, and therapy tasks.
Education