Meril
Website:
merillife.com
Job details:
Role Overview
We are looking for a Senior Agentic AI / RAG Engineer to build intelligent robotic systems powered by LLMs, Agentic AI, and real-world perception pipelines. This role focuses on enabling robots to reason, plan, and act autonomously by combining RAG-based knowledge systems with robotics frameworks (ROS, sensors, control systems).
You will bridge AI cognition (LLMs) with physical execution (robots).
Key Responsibilities
- Design agentic AI architectures for robotics (planning, reasoning, tool usage in real-world tasks)
- Build RAG pipelines to enable robots to query knowledge (manuals, environment data, SOPs)
- Integrate LLMs with robotics stacks (ROS/ROS2) for decision-making and task planning
- Develop systems for:
- Task decomposition
- Autonomous navigation & decision making
- Human-robot interaction (natural language interfaces)
- Connect AI agents with robot hardware APIs, sensors, and actuators
- Work on real-time constraints, edge deployment, and latency optimization
- Implement vision + language models (VLMs) for perception-driven reasoning
- Ensure safety, reliability, and fallback mechanisms in autonomous systems
Required Skills & Qualifications
- 5+ years in AI/ML, Robotics, or Software Engineering
- Strong programming in Python (C++ is a big plus for robotics)
- Hands-on experience with:
- ROS / ROS2
- Robotics simulation tools (Gazebo, Isaac Sim, Webots, etc.)
- Sensor integration (LiDAR, cameras, IMU)
- Experience with:
- RAG pipelines & vector databases
- LLM frameworks (LangChain, LlamaIndex, etc.)
- Agent-based systems (planning + tool use)
- Understanding of:
- SLAM, path planning, and control systems
- Computer vision (OpenCV, deep learning models)
- Real-time system constraints
Preferred Qualifications
- Experience with embodied AI / autonomous robots
- Exposure to multi-agent robotics systems
- Experience with edge AI deployment (NVIDIA Jetson, embedded systems)
- Familiarity with VLMs (Vision-Language Models) and multimodal AI
- Experience integrating LLMs with physical systems
- Candidates based in or willing to relocate to Gujarat preferred
Soft Skills
- Strong systems thinking (hardware + software integration)
- Ability to debug complex real-world systems
- Ownership mindset and experimentation-driven approach
What You’ll Build
- Robots that can understand instructions in natural language
- Systems that reason before acting in physical environments
- AI agents that combine memory (RAG) + perception + action
Click on Apply to know more.