
1
1
United Arab Emirates
4+ years of experience
I have several years of experience in computer vision for drones, including tree counting, forest mapping, and agricultural monitoring. I’ve worked on detection pipelines, dataset preparation, and real-time inference for embedded hardware. I also develop interactive educational games for children, combining vision models, gesture/sound recognition, and playful UX to keep kids engaged while learning. I love building meaningful AI products that can help people in real life — and I’d be happy to team up with anyone passionate about computer vision, healthtech, or impactful technology.

TaskDigest addresses a real operational pain: managers and teams are overwhelmed by fragmented information spread across Slack threads, emails, and daily standups. Important risks, blockers, and follow-ups are easily missed in the noise. TaskDigest is an LLM-centered system that transforms unstructured operational updates into a clear daily operations digest. The LLM first extracts meaning from raw human language - identifying tasks, risks, blockers, and decisions across multiple channels. Deterministic, rule-based logic then applies transparent policies to ensure nothing critical is lost. Finally, the LLM translates these structured results back into a concise, human-readable report. The system is designed for trust and production use: every output is traceable from raw input to final summary and it runs locally on company infrastructure, so sensitive data never leaves the organization. TaskDigest helps managers stop drowning in messages and start acting immediately.
7 Feb 2026