Abdul Arham Khan

AI-Backend Engineer · Clinical history & compliance · Part-time

Lahore, Punjab, Pakistan · On-site

• Design and build multi-tenant platforms for HIPAA/PHI-compliant clinical history management, with end-to-end ownership of architecture aligned to US healthcare data-security expectations.
• Ship AI-assisted workflows that scribe prescriptions, reports, and clinical histories into production, integrated safely with the rest of the stack.
• Develop backend services in Node.js and Python; build and maintain RESTful APIs and connect AI and third-party services into secure, auditable release paths.
• Implement authentication, authorization, and secrets management; harden for reliability and performance as usage grows.
• Partner with frontend and ML teams to deliver features end-to-end, from design through deployment and operations.

AI/ML Engineer · Conversational AI and voice systems

Lahore, Punjab, Pakistan · On-site

• Engineer AI voice agents and deploy production-grade systems optimized for low-latency, real-time interactions.
• Design and implement advanced barge-in handling and ASR/IVR optimizations to manage high-volume patient scheduling.
• Develop FastAPI endpoints for telephony workflows and integrate streaming STT/TTS (Deepgram, Cartesia) with LLM responses.
• ntegrate telephony platforms (Twilio, Vonage, Pipecat, Retell) and manage secure deployments using Docker and CI/CD on VPS.
• Optimize audio pipelines, reduce latency, and improve recognition accuracy with retry/backoff and graceful degradation strategies.
• Collaborate with product and backend teams to design scalable APIs and monitoring for uptime, latency, and call-quality metrics.

Backend Engineer (Contract) · AI-first Software Engineer (Part-time)

Remote · Chicago, Illinois

• Built and maintained backend services for web/mobile applications with high scalability requirements.
• Designed and optimized data structures and backend APIs, including 50+ endpoints for With a Twist.
• Collaborated across product, design, and engineering to deliver dependable AI-powered features.

Frontend development intern

Remote · Delaware, US

• Job board application with TypeScript, React, and Tailwind CSS; enhanced UI with Chakra UI, Ant Design, and Material UI.
• Supabase as BaaS for a solid, efficient application foundation.

TypeScript, Next.js, Node.js, PostgreSQL, Prisma, Redis, Docker, Kafka, Digital Ocean

Labyrinth is a web-based collaboration platform designed to connect students, freelancers, and professionals through streamlined matchmaking and workspace tools. Built 84+ REST API endpoints and completed SDLC with CI/CD in 6 months. Modules included recommendation system, chats, and project management.

Python, FastAPI, Twilio, GPT-4o-mini, Deepgram STT/TTS, Pipecat, Vonage, Retell, Hostinger

Built and deployed medical voice agents optimized for low-latency interaction; implemented advanced barge-in handling and ASR/IVR optimizations to manage high-volume patient scheduling seamlessly.

TypeScript, React Native, Node.js, PostgreSQL, Prisma, Redis, Docker, Digital Ocean

An AI-first mobile game for playing puzzles with P2P, P2C, and multiplayer game rooms, plus puzzle creation using RAG, Ollama, and a vector DB. Efficient online and offline features with social circle and leaderboards. Robust RESTful API endpoints on WebSockets. Redis minimizes request times to sub-milliseconds. Redis Pub/Sub event-driven architecture for notifications.

TypeScript, React Native, Expo, Convex

Payer is a modern, secure mobile wallet that simplifies digital transactions. Send money instantly to friends and family, recharge mobile balances for all major Pakistani carriers, and manage finances with ease. Designed pixel-perfect UIs, developed transactional CRUD with robust middleware and error handling, and deployed to the cloud.

JavaScript, Node.js, AWS SES, S3, KMS, PostgreSQL

PAWS is a HIPAA and SOC 2 compliant platform for US clinical organizations, built at Quishub. Engineered HIPAA-compliant, multi-tenant clinical platforms with AI-based scribing; led architecture and compliance. Built secure Node.js/Python APIs and integrated AI services into production. Strengthened auth, secrets, and system reliability while delivering features with frontend and ML teams.

Python, TensorFlow, MLflow, MobileNet, Streamlit, Docker, Git CI/CD, Kaggle

Built a continuous integration and delivery pipeline for model updates to automate retraining with new datasets and monitoring model performance over time.