AI/ML Engineer and Software Developer focused on building production-ready intelligent applications using Machine Learning, Deep Learning, Computer Vision, Python, and modern software engineering technologies.

My primary interests are Artificial Intelligence, Machine Learning, Deep Learning, Computer Vision, Generative AI, and AI application development.

I enjoy taking AI projects beyond model training — from data preprocessing and experimentation to backend APIs, user interfaces, deployment, and complete production applications.

My AI/ML experience includes:

• Machine Learning and Deep Learning

• Computer Vision and Image Classification

• Transfer Learning and Ensemble Learning

• Explainable AI using Grad-CAM

• TensorFlow and model development

• Data preprocessing and augmentation

• Model evaluation and optimization

• Python-based AI application development

• FastAPI model-serving APIs

• AI model integration with web applications

One of my key projects is SafeSkin AI, a deep-learning-based skin cancer detection platform built using the HAM10000 dataset. I worked with DenseNet201 and InceptionV3 transfer-learning models, ensemble learning, class-imbalance handling, image augmentation, Grad-CAM explainability, FastAPI, and a full web-based inference workflow.

Alongside AI/ML, I have hands-on software engineering and full-stack development experience using:

• Python

• C/C++

• JavaScript

• TypeScript

• React

• Next.js

• Node.js

• FastAPI

• PostgreSQL

• MongoDB

• REST APIs

• Docker

• Git/GitHub

• Vercel

I have built production-style applications involving authentication, role-based access control, APIs, databases, payment integrations, admin dashboards, cloud storage, automation, and deployment.

I also have experience working with IoT and embedded systems, giving me additional understanding of how software interacts with real-world devices and hardware.

My target opportunities:

• AI/ML Engineer

• Machine Learning Engineer

• AI Engineer

• Deep Learning / Computer Vision Engineer

• Software Development Engineer

• Full-Stack Software Engineer

I am especially interested in teams building practical AI products where machine learning, backend engineering, APIs, and modern application development work together.

Jun 2026 – Present · Full-time (Campus Placement)

Working as a Software Development Engineer with responsibilities across software engineering, Python development, IoT systems, embedded integration, automation, and product development.

Key responsibilities and work:

• Develop software and tools using Python, C/C++, and modern development technologies.

• Build Python-based automation, debugging, testing, and development utilities.

• Design and integrate APIs and application components for hardware-connected products.

• Work on communication between software applications and embedded devices.

• Develop and debug STM32 and Arduino-based systems.

• Work with UART, SPI, I²C, USB, and device communication protocols.

• Analyze hardware interfaces, evaluation boards, sensors, and PCB-level communication.

• Contribute across firmware, backend services, application software, testing, and deployment.

• Research AI/ML integration opportunities for intelligent software and connected-device products.

• Work across complete product-development lifecycle, from prototype and debugging to application integration.

Dec 2025 – Jun 2026 · Patna, Bihar

• 📊 Digitised and restructured 10,000+ legacy land records into a queryable relational schema
• ✅ Achieved 100% data integrity across migrated records with validation scripts
• ⚡ Improved data retrieval speed by an estimated 50% across 11 Block-level workflows in Banka District

Jul 2023 – Jun 2025 · 4 engagements across Bihar

• 🎯 90%+ project completion rate among 300+ students
• 🏫 Taught at SVN Infra & Solar Services, TechGlaz Labs, NKB Infinity
• 🔧 Designed structured lab modules in Arduino programming, sensor integration, IoT prototyping

A production-grade Java SDK for controlling 124 Broadlink IoT device types across 19 categories — IR remotes, smart plugs, sensors, bulbs, thermostats, curtains, alarms, and hubs.

📱 Android · ☕ Spring Boot · 🖥️ Desktop Java · 🔷 Kotlin · ⚡ Zero Dependencies

• broadlink-core — ZERO external dependencies, runs on JDK 11+
• SPI pluggable architecture (Logger, JsonSerializer, NetworkProvider)
• Reverse-engineered proprietary UDP protocol with AES-128-CBC encryption
• 53 unit tests validating exact packet byte structures against reference implementation

BroadlinkClient client = new BroadlinkClient();
List<BroadlinkDevice> devices = client.discover().get();
devices.get(0).authenticate().get();

if (device instanceof RmMiniDevice rm) {
    rm.enterLearning().get();
    byte[] irCode = rm.checkData().get();
    rm.sendData(irCode).get();
}

• 💼 LinkedIn: linkedin.com/in/sushant-sagar-723b65338
• 💻 GitHub: github.com/Sushantku1099
• 📧 Email: sushant1892004@gmail.com
• 📞 Phone: +91-6207504908
• 📍 Location: Pawai Dih, Banka, Bihar — 813101

Open to opportunities in Software Development, AI/ML Engineering, IoT & Embedded Systems, and Full Stack Development.