Job Description: IoT & AI Developer
Location: Mumbai / Hybrid
Employment Type: Full-Time
Experience: 4+ Years
About the Role
We are looking for a hands-on IoT & AI Developer who can bridge the gap between embedded/edge hardware, IoT systems, and modern Artificial Intelligence.
The ideal candidate will have experience developing solutions using Arduino, Raspberry Pi, sensors, embedded systems, and IoT communication protocols, combined with strong expertise in local AI models, model inference, fine-tuning, training, quantization, distillation, and deployment into secure private AI environments.
This role is particularly suited to an engineer who enjoys taking an idea from sensor/device → edge processing → AI model → private inference infrastructure → production application.
Key Responsibilities IoT & Edge Development Design and develop IoT solutions using Arduino, Raspberry Pi, ESP32, and similar embedded/edge computing platforms. Interface with sensors, actuators, cameras, industrial equipment, and other peripheral devices. Develop embedded and edge applications using Python, C, and C++.
Implement device communication using protocols such as: MQTT HTTP/HTTPS WebSockets REST APIs Modbus CAN UART SPI I2C BLE Wi-Fi Ethernet Build reliable device-to-edge and edge-to-server communication architectures. Develop solutions that can operate in offline, intermittent-connectivity, and fully private network environments. Implement local data acquisition, filtering, preprocessing, caching, and event processing. Optimize applications for resource-constrained edge hardware. Edge AI & Local AI Models Deploy and operate AI models locally without dependence on public cloud AI services. Implement AI inference on Raspberry Pi, ARM devices, edge computers, workstations, GPUs, and private servers. Work with local and open-weight AI ecosystems including technologies such as: PyTorch TensorFlow ONNX / ONNX Runtime TensorFlow Lite llama.cpp Ollama MLX Hugging Face Transformers OpenVINO NVIDIA TensorRT NVIDIA Jetson Optimize models for CPU, GPU, NPU, and other edge accelerators. Benchmark models for: Latency Throughput Memory usage Power consumption Accuracy Model size AI Model Training & Fine-Tuning Prepare datasets for model training and fine-tuning. Perform data cleaning, labeling, preprocessing, augmentation, and validation. Train and fine-tune machine learning, deep learning, computer vision, and language models. Work with techniques such as: Transfer learning Supervised fine-tuning Parameter-efficient fine-tuning LoRA / QLoRA Prompt tuning Embedding models Retrieval-Augmented Generation Evaluate trained models using appropriate performance and quality metrics. Create reproducible AI experimentation and model-development pipelines. Model Distillation & Optimization Distill larger AI models into smaller models suitable for local and edge deployment. Implement and evaluate teacher-student model architectures. Apply model optimization techniques including: Knowledge distillation Quantization Pruning Weight compression Mixed precision Operator optimization Model conversion Work with formats and quantization technologies such as: GGUF ONNX FP16 / BF16 INT8 INT4 Balance model accuracy against inference speed, memory requirements, and hardware limitations. Private AI Deployment Design and implement fully private/on-premises AI infrastructure where data and models remain within the customer's environment. Deploy local AI models on: Private servers GPU servers Workstations Edge gateways Raspberry Pi devices NVIDIA Jetson devices Private Kubernetes environments Build internal inference APIs and AI microservices. Package AI applications and inference services using Docker containers. Implement scalable model-serving solutions using suitable inference engines. Integrate AI inference with existing enterprise applications and IoT platforms. Support environments with no Internet connectivity or air-gapped infrastructure. Generative AI & Local LLM Development Deploy and integrate local Large Language Models and multimodal models. Build private AI applications using: Local LLMs Vision Language Models Embedding models Speech models Computer Vision models Develop RAG-based applications using private organizational datasets. Implement vector search and semantic retrieval. Work with vector databases such as: PostgreSQL / pgvector Qdrant Milvus Weaviate Chroma Develop local AI agents and intelligent automation workflows where appropriate. Optimize LLM context size, inference parameters, memory utilization, and response latency. Computer Vision & Sensor AI Experience with one or more of the following would be highly desirable: Object detection Object tracking Image classification Facial/object recognition Anomaly detection OCR Video analytics Sensor fusion Time-series analysis Predictive maintenance Audio/speech processing Environmental sensor analytics Industrial equipment monitoring Backend & Integration Development Develop backend services and APIs for IoT and AI applications. Strong programming experience with Python. Experience with C/C++ for embedded or performance-sensitive applications. Experience developing REST APIs using frameworks such as: FastAPI Flask Django Work with databases such as: PostgreSQL SQLite TimescaleDB InfluxDB Redis Develop event-driven architectures using MQTT or message brokers. Integrate devices and AI systems with web, mobile, and enterprise applications. Infrastructure & DevOps Containerize IoT backend and AI workloads using Docker. Experience with Docker Compose and preferably Kubernetes. Deploy applications on Linux-based systems. Configure GPU-enabled containers where required. Maintain model versions, deployment configurations, and inference environments. Implement CI/CD pipelines for application and AI model deployments. Monitor device health, service availability, model performance, and infrastructure utilization. Security & Privacy Design IoT and AI architectures with security and privacy as core requirements. Secure device-to-server communication using TLS and appropriate authentication mechanisms. Implement secure provisioning and device identity management. Protect sensitive IoT data, model artifacts, credentials, and AI datasets. Build AI solutions where confidential information remains within private infrastructure. Understand security requirements associated with edge devices and potentially disconnected environments. Required Technical Skills The candidate should have strong experience with several of the following: Programming Python C C++ Shell scripting IoT / Embedded Arduino Raspberry Pi ESP32 Linux-based edge devices Sensors and actuators MQTT Modbus BLE Serial communication GPIO / SPI / I2C / UART AI / Machine Learning PyTorch TensorFlow Hugging Face Transformers ONNX TensorFlow Lite Model inference Model training Fine-tuning Model distillation Quantization Computer Vision Generative AI / LLMs Local AI Ollama llama.cpp GGUF ONNX Runtime MLX OpenVINO TensorRT NVIDIA CUDA Infrastructure Linux Docker Docker Compose Git CI/CD REST APIs
Preferred Qualifications :
Bachelor's or Master's degree in Computer Science, Electronics, Embedded Systems, Artificial Intelligence, Robotics, Electrical Engineering, or a related discipline.
4+ years of software, IoT, embedded systems, machine learning, or AI development experience.
Demonstrable experience building physical IoT prototypes and taking them into production.
Experience deploying AI models on hardware with limited CPU, RAM, storage, or power. Experience developing offline-first or completely on-premises AI systems.
Understanding of model licensing and commercial usage considerations for open source/open-weight AI models. Experience working with GPU infrastructure and NVIDIA CUDA is desirable.
Experience with NVIDIA Jetson or other edge AI accelerators is a strong advantage.
What We Are Looking For We are particularly interested in candidates
who can demonstrate end-to-end ownership.
A strong candidate should be capable of taking a requirement such as: Sensor / Camera → Arduino or Raspberry Pi → Data Processing → AI Model → Edge Inference → Private AI Server → Application / Dashboard and independently designing, prototyping, optimizing, and deploying the complete solution.
The individual should be comfortable working across hardware and software boundaries and should have a strong experimentation mindset when evaluating new AI models, inference engines, edge accelerators, and deployment approaches.
Key Competencies Strong problem-solving and debugging skills.
Hands-on approach to hardware and software development.
Ability to rapidly prototype new IoT and AI concepts.
Strong understanding of AI model performance and optimization. Ability to evaluate new open-source AI technologies.
Ability to work independently on complex R&D initiatives. Strong documentation and communication skills.
Ability to transform research prototypes into reliable production systems.
Nice-to-Have Experience Experience in any of the following would be an additional advantage: Robotics Industrial IoT / IIoT Predictive maintenance Digital twins Edge computing Drone/UAV systems Autonomous systems NVIDIA Jetson Coral TPU Intel NPU/OpenVINO Apple Silicon / MLX Private AI clusters Local multimodal AI Speech-to-text / text-to-speech RAG and vector databases AI agents Kubernetes-based inference MLOps / LLMOps Model monitoring and evaluation Air-gapped AI deployments Hardware acceleration and inference benchmarking.
''If interested, please share your updated CV at [email protected]''
Pay: Up to ₹3,000,000.00 per year
Benefits:
- Cell phone reimbursement
- Flexible schedule
- Food provided
- Health insurance
- Internet reimbursement
- Life insurance
- Paid sick time
- Paid time off
- Provident Fund
Work Location: In person
