I build systems where AI, software, machines, sensors, and real operational workflows have to work together reliably.
Currently a Visiting Researcher at EDGE BRIDGE, Chief Product Officer at Kanban Studios, and Computer Science student at Al Ain University.
My work sits mainly across AI systems, robotics, industrial automation, Digital Twins, and reliability engineering.
- Building AI systems that reason from evidence, system state, and policy.
- Connecting software with robots, machines, sensors, and operators.
- Developing Digital Twins and HMIs for monitoring and control.
- Testing whether systems remain reliable when conditions actually change.
Agents, decision-support tools, RAG pipelines, knowledge systems, and computer vision.
ROS integrations, Digital Twins, HMIs, RFID workflows, machine interfaces, and edge systems.
APIs, internal tools, dashboards, monitoring systems, automated testing, and production-focused prototypes.
- Start with the real workflow and failure modes, not the feature list.
- Treat AI as part of a larger system, not the entire product.
- Keep humans involved when automated decisions have consequences.
- Measure performance instead of assuming something works.
- Build systems that are explainable, observable, testable, and recoverable.
RAG Qdrant Knowledge Graphs MCP Agentic AI
RFID Node-RED HMI Digital Twins ASRS Computer Vision
OpenTelemetry SigNoz k6 Load Testing
An AI reliability agent that determines whether a code change is actually safe before production.
It compares old and new versions under identical load, investigates traces and logs through SigNoz, identifies regressions, proposes focused fixes, and verifies them in isolation.
Detected one failure where 94.55% of requests were breaking, and another where errors remained at 0% while P95 latency jumped from 103.91 ms to 2,376.92 ms.
FastAPI Next.js OpenTelemetry SigNoz MCP k6
2nd Place Nationally — AIMed AI Builders Summit 2026
A surgical operations platform that detects cases at risk of cancellation, identifies readiness blockers, and helps staff recover operating-room capacity through human-approved decision support.
AI Healthcare Decision Support Human-in-the-Loop
An automated infrastructure-testing system that finds the smallest configuration capable of meeting defined performance targets.
One experiment improved P95 latency from 496 ms to 265 ms. Another detected that adding a container made performance roughly 2.5× worse and automatically rejected the configuration.
Next.js Go PostgreSQL TypeScript autocannon
What happens when information an AI previously learned becomes wrong?
Aletheia detects conflicting or retracted information, updates its knowledge graph, and regenerates answers from trusted evidence.
Instead of only learning more information, it can also unlearn invalidated knowledge.
Knowledge Graphs AI Memory Evidence Tracking AI Agents
A webcam-controlled spatial canvas with gesture input, editable vector strokes, and depth-aware interaction for hands-free drawing.
Python MediaPipe OpenCV
A touch-focused industrial HMI for controlling a simulated robotic arm and conveyor system with live position, movement, direction, and machine-state feedback.
Svelte HMI Digital Twin
Jan 2026 – Present
Applied R&D across industrial automation, Digital Twins, HMIs, robotics, computer vision, and intelligent machine systems.
Jun 2026 – Present
Product direction and technical planning for AI products, automation systems, internal tools, and intelligent workflows.
Jul 2026
Led project management and technical integration for a multi-member Flutter development team across architecture, GitHub workflow, integration, QA, and release preparation.
Sep 2025 – Dec 2025
Worked on ASRS, RFID/backend integration, ROS systems, Digital Twins, hardware commissioning, computer vision, and system reliability.
Contributed to resolving 50+ tracked software and system issues.
Aug 2025 – Sep 2025
Worked on robotics, industrial automation, Boston Dynamics Spot exploration, ASRS planning, and hardware-software integration.
| Result | Competition / Event | Project |
|---|---|---|
| 1st Place | Tamkeen 5.0 Hackathon by G42 | Agentic Refund Intelligence |
| 1st Place | Replit × IEC Buildathon | Starkz AI |
| 2nd Place Nationally | AIMed AI Builders Summit 2026 | CaseReady AI |
| 2nd Place | Safe AI Cup 2026 | Generative AI / Education |
| 2nd Place | Emirates Agriculture Hackathon | SooqRoot |
| 3rd Place | c0mpiled-2 UAE Hackathon | DeedFlow |
| 6th / 26 Teams | Abu Dhabi AI PropTech Challenge | Reach |
| Finalist | Global Build Challenge | Common Ground |
| Finalist | EDGE LIF NextGen Industry Challenge | FalconPatrol AI |
| Top 170 / 4,710 | Create Apps Championship | XPBridge |
A few of these projects explored AI safety, healthcare operations, agriculture, real estate, robotics, and urban intelligence.
At EDGE BRIDGE, I have worked directly with:
ASRS HMI Digital Twins RFID ROS Machine Integration Computer Vision mTLS Hardware Commissioning
I also presented EDGE BRIDGE automation systems at Make it in the Emirates 2026, demonstrating ASRS, HMI, and Digital Twin capabilities to visitors and industry stakeholders.
BSc Computer Science · 2023 – Present
Outside of tech, I play football and enjoy drawing and painting.
I am interested in building intelligent systems that have to survive contact with the real world.
LinkedIn · Email · Kanban Studios · GitHub


