Gray Scrubs AI Aug 2025 – Present

Founding Engineer, Computer Vision & Machine Learning, operating-room computer vision startup, NVIDIA Inception · London, UK (Remote)

• Built an RGB-only, role-labelled 3D digital twin of the operating room from five ceiling cameras: fine-tuned SAM 3 (848M) for OR panoptic segmentation on one GPU, patching its upstream training code, and fused it with feed-forward multi-view reconstruction. Benchmarked it against 7 published 3D methods on depth-sensor ground truth; an ablation showed the domain fine-tune carries the semantics.
• Fine-tuned LTX-Video 13B (LoRA) on 645 multi-camera OR clips for mask- and depth-controlled video generation. Now simulating physics-labelled rare-event training video in Isaac Sim, and reproduced the OmniFall fall-detection baseline on AWS GPUs and the UK's Isambard-AI supercomputer (GH200, Slurm), for a CVPR 2027 paper in preparation.
• Designed an LLM labelling pipeline (LLM labellers, blind judges from two model families, human review in self-hosted CVAT) that produced 6.7K hand-contact labels; fine-tuned the HOI-DETR hand-object detector on them and extended it into an RGB-only sterile-field touch detector, configured on validation before one held-out test read.
• Earlier experiments: streaming surgical-phase recognition (frozen DINOv3 and V-JEPA 2 features, causal HMM decoding), PaliGemma 2 vision-language QLoRA for procedure coaching, and a multitask model exported to ONNX for Jetson.
• Built the clinical data side on AWS (Terraform): a hospital video-ingestion platform (recorder-disk reader, resumable S3 upload, GPU screening workers, a passkey-protected FastAPI/React review portal) and DICOM de-identification (OCR quarantine of burned-in text, NER scrubbing of reports, CT defacing) with Lambda/S3 portals for surgeons and data partners.

Arantic Digital Oct 2025 – Present

Full-Stack Software Engineer (Part-time), software agency, team of 4–6 engineers · Munich, Germany (Remote)

• Primary engineer of a generative-AI marketing content app in production since Jul 2026: LLM generations run as BullMQ jobs with live SSE progress, timeouts and stalled-job recovery; thread history with refine and fork in 20 languages; file attachments passed to the model as context.
• Wrote the provider-agnostic LLM layer (Vercel AI SDK; Anthropic, OpenAI, Mistral) with a live model catalog, cost estimates and encrypted keys, plus a prompt editor whose layouts compile to Zod schemas for structured output.
• Designed the real-time notification system: Server-Sent Events fed by a BullMQ worker in its own container, with cross-tab leader election so each user holds one connection; later merged with AI progress into one multiplexed event stream.
• Implemented a recursive-descent search query parser and SQL builder over PostgreSQL full-text search (weighted tsvector, GIN), and a product module with trigram search and a reusable transactional CSV import.

Contact Center Specialists Jul 2024 – Aug 2025

Backend Developer, Infrastructure · Cheyenne, WY (Remote contract)

• Ran production infrastructure (FreeSWITCH telephony, PostgreSQL, MongoDB, Redis, Zabbix and Grafana monitoring, call-record billing); parallelised the Go device-provisioning service and built a real-time voice-AI flow on Gemini Live.

SCIT CAIR Jan 2024 – Jun 2025

AI Researcher · Lahore (Hybrid, part-time)

• Remote-sensing research with Dr. Usman Nazir: ViTs and CLIP filtering on 600K+ satellite images (NeurIPS 2024 paper).

Beaconhouse National University Dec 2024 – May 2025

Lead Full-Stack Developer · Lahore (Hybrid, part-time)

• Built a Next.js/tRPC LMS with role-based access and SSO, and a geospatial analytics dashboard for HBL Bank.

VoiceFlow Released

Python, faster-whisper, CUDA, Pyloid, React

Local voice dictation and meeting recorder for Windows and Linux. Hold a hotkey to dictate, or record long meetings with system audio.

• Whisper transcription on the local GPU
• Meeting mode captures mic and system audio, with bring-your-own-LLM summaries
• Open source

github.com/infiniV/VoiceFlow

Android-Ui-MCP Released

TypeScript, MCP, adb, Docker

MCP server that lets AI coding agents see a running Android app: screenshots from a device or emulator while you build with Expo, React Native, Flutter or native Android.

• Two tools: take_android_screenshot and list_android_devices
• Works with Claude Code, Claude Desktop, GitHub Copilot and Gemini CLI
• Published on npm as android-ui-assist-mcp

github.com/infiniV/Android-Ui-MCP

Dori Released

Shell, QML, adb, scrcpy, v4l2loopback

Omarchy plugin that turns an Android phone into a webcam and a second screen on Linux, over USB or Wi-Fi. Meet, Zoom, Teams, Discord and OBS see an ordinary webcam.

• Phone camera as /dev/video10, with rotation, resolution and frame rate on the panel
• Screen mirror, screenshots to the clipboard, background recording and phone notifications
• No app and no account on the phone

github.com/infiniV/omarchy-dori

claude-usage-waybar Released

Python, GTK4, libadwaita, Waybar

Waybar module for Claude Code: five-hour and seven-day rate limits on the bar, today's tokens and spend, and live session stats in a GTK4 popover.

• Reads the rate_limits block from Claude Code's statusline
• Today's tokens and dollar cost via ccusage
• Follows the active Omarchy theme

github.com/infiniV/claude-usage-waybar

Representation or readout? Model stitching locates different captioning failures in π₀ and π₀.₅ Submitting

UniReps 2026 (NeurIPS workshop)

R. Arbaz, et al.

Interpretability of robot vision-language-action models: model stitching, probes and sparse crosscoders in pre-registered experiments on sealed test sets.

Mapping Air Pollution Sources with Sequential Transformer Chaining: A Case Study in South Asia Published

NeurIPS 2024 Climate Change AI Workshop

H. M. Abubakar, R. Arbaz, H. Ahmad, M. Nazir, U. Nazir

Pollution-source detection (factory and brick-kiln chimneys) in South Asian cities from satellite imagery: a Vision Transformer filters high-resolution tiles, then Remote CLIP refines detection in two passes.

PDF arXiv 2410.19629 Workshop page

Beaconhouse National University Aug 2021 – Jun 2025

BSc (Hons) Computer Science · Lahore

Dean's Honour List 2024 and 2025