Mohammad
Rajaei
Senior Software Engineer with 6+ years of experience designing scalable software across AI, robotics, distributed systems, and modern web platforms. Experienced in leading architecture, modernizing engineering systems, and delivering production-ready solutions.
Building systems that matter
I'm a Senior Software Engineer based in Tehran, Iran, with a strong passion for solving complex engineering challenges at the intersection of AI, distributed systems, and modern web platforms.
Over the past 6+ years, I've led the development of end-to-end AI pipelines for medical imaging, modernized shipment visibility platforms used globally, and engineered mission-critical robotic automation systems for pharmaceutical manufacturing.
I thrive in environments where engineering rigor meets product thinking — driving architectural decisions, mentoring teams, and shipping production-ready solutions that create measurable business impact.
Currently pursuing a Master's in Artificial Intelligence at Azad University (IAU), deepening my expertise in ML systems and computer vision.
Technical Expertise
A comprehensive stack spanning AI research, full-stack web development, distributed systems, and embedded hardware engineering.
AI & Computer Vision
Agentic AI & LLM
Frontend
Backend
Databases
Architecture
Cloud & DevOps
Observability & Tools
Hardware & IoT
Work History
6+ years across AI startups, logistics tech, pharmaceutical robotics, and consumer platforms.
- Achieved 98.5% Precision, 97.8% Recall, 99.2% mAP@0.5, and 189 FPS through iterative optimization of AI pipelines.
- Integrated and standardized two large-scale medical cytogenetics datasets containing 250,000+ chromosome instances from 1,600+ prenatal cases into a unified COCO format.
- Architected asynchronous FastAPI services enabling reliable long-running AI inference jobs.
- Fine-tuned state-of-the-art models including YOLO, RF-DETR, SAM2, and MaskNet for production.
- Reduced hypothesis validation time to one week by building rapid MVPs with Gradio and Next.js.
- Contributed to recruitment and onboarding, reducing ramp-up time by 40%.
- Improved LCP (Largest Contentful Paint) by 30% and boosted overall page speed by 15% through Core Web Vitals optimization.
- Drove a 50% increase in sales within the first quarter post-launch by partnering with UI/UX and Product teams.
- Implemented unified frontend and backend error monitoring with Sentry enabling end-to-end tracing.
- Engineered n8n workflow automation saving approximately 16 engineering and business hours per month.
- Designed Grafana dashboards to monitor API performance, latency, throughput, and production failures.
- Developed and maintained the main Dashboard using React.js, GraphQL, and Apollo for global shipment visibility.
- Scaled production capacity from 12,000 to 46,000 units via distributed master-slave robotic control using async Python, Raspberry Pi, MongoDB, and Arduino.
- Developed a real-time HMI with React enabling operators to monitor robot status and control workflows.
- Fine-tuned PyTorch computer vision models for visual inspection accuracy in robotic manufacturing.
- Programmed embedded firmware for Arduino Due controlling motors, drivers, optical and mechanical sensors.
- Engineered runtime health monitoring, fault detection, and recovery mechanisms preventing equipment damage.
- Led and mentored a team of 4 frontend engineers, reducing production bugs by ~60% and successfully mentoring 3 engineers to promotion within 8 months.
- Orchestrated full migration of Vidaneh's frontend to Next.js and TypeScript establishing modern development patterns.
- Deployed a real-time AI-powered logo detection system to identify and hide logos in videos per platform policies.
- Architected a modular advertising platform supporting banner, audio, and video ads.
- Created an interactive SVG-based web game to enhance user engagement and increase session duration.
- Resolved a critical database deadlock affecting a medical content platform serving ~40,000 monthly visitors through root-cause analysis and indexing optimization.
- Led full-stack development of an on-demand mechanic platform with real-time technician tracking using React and Laravel.
- Designed and developed a personalized diet planning application with custom meal planning and health management features.
- Architected a B2B vendor networking platform connecting suppliers with retailers through product catalogs and digital marketing tools.
- Delivered multiple e-commerce platforms and customer loyalty systems using Laravel, React, WordPress, and WooCommerce.
Open Source & Hobby
Side projects I build to explore ideas, teach concepts, and sharpen tools I care about — from protocols and databases to visual computer vision workflows.
Build computer vision pipelines visually — inspect every pixel transformation in real time, then export clean, production-ready Python the moment you're happy with the result. Assemble your pipeline as a graph, watch every node's output update live, and walk away with production code — not a black box.
- Visual graph editor for computer vision pipelines
- Live per-node output inspection while you tune parameters
- Breaks the tweak → run → save → squint debug loop
- Export production-ready Python when the pipeline looks right
Wireflow
Interactive network protocol playground
Learn how REST, streaming, SSE, WebSocket, gRPC, polling, GraphQL, WebRTC, MQTT, and more move data between clients and servers. Everything runs in the browser — step animations, failure scenarios, side-by-side compare, and a short chooser quiz. No accounts or live traffic.
- Step simulator with play/pause, speed control, and looping packet motion
- Failure scenarios: timeouts, reconnects, rate limits, ICE failure, and more
- Comparison table and dual-protocol compare derived from protocol traits
- Protocol chooser quiz — three questions → recommended protocol
- Shareable deep links like #websocket/happy/3 or #rest/rate-limit/1
DataFlow
Interactive database learning path
Start with SQL fundamentals, then internals and families, and finish able to choose a storage architecture for workload and guarantees — and grow from a handful of requests to billions. Everything runs in the browser: leveled lessons, Mermaid diagrams, concept simulators, a workload chooser quiz, side-by-side compare, shareable deep links, and local progress. No accounts required.
- Seven levels — Beginner → Intermediate → Advanced → Distributed → Data architecture → DB engineering → Architect
- Lesson reader with prose, SQL examples, ASCII + Mermaid diagrams, and takeaways
- Concept simulators for Postgres path, transactions, replication, polyglot, and more data models
- Database chooser starting with “why not PostgreSQL?”, then adaptive questions → engine + checklist
- Comparison matrix, localStorage progress, quick find (Ctrl/⌘ K), and deep links like #beginner/joins
Education
Master's Degree
Artificial Intelligence
Azad University (IAU)
Bachelor's Degree
Computer Software Engineering
Azad University (IAU)
Certifications
Agentic AI
Maktabkhoone.org · 2026
OpenCV
Maktabkhoone.org · 2026
Big Data
7Learn.com · 2026
Feedback Workshop (HRM)
Diginext · 2023
Engineering Management & Software Architecture
Diginext · 2023
Generative Adversarial Networks (GAN)
Diginext · 2022
Machine Learning
Diginext · 2022
Let's work together
I'm currently open to new opportunities — remote positions, relocation, or exciting technical challenges in AI, distributed systems, or full-stack engineering. Feel free to reach out and I'll get back to you as soon as possible.
Send an Email