New opportunity

Robotics Engineer

Xenonstack Private Limited · Mohali, India

About this role

XenonStack seeks a Robotics Engineer in Mohali, India, to develop embodied agents and humanoid and autonomous robots from concept through production. The role combines mechanical systems, embedded software and AI for potential applications in public safety, surveillance, manufacturing and healthcare. The posting’s Job Information section lists 1–3 years of work experience, while the opportunity description seeks 2–5 years in robotics, AI integration and mechatronics. The listed salary is 5 - 7 LPA.

Responsibilities

  • Design, build and program robotic systems that integrate Vision AI, sensors and embedded controllers. Develop control algorithms for navigation, motion planning and object manipulation.
  • Integrate agentic AI workflows for autonomous decisions; connect vision models for object detection, tracking and anomaly detection to robotic control loops; and use LLMs for multimodal perception and reasoning.
  • Prototype and test systems using robotics simulation platforms. Validate performance in real-world conditions and support reliable, safe and compliant deployments.
  • Collaborate with Platform and AgentOps teams on field deployment, optimize software for edge hardware such as Jetson, Raspberry Pi, FPGAs and custom boards, and implement feedback loops to improve robotic capabilities.

Requirements

  • Strong Python, C++ and ROS proficiency; experience with computer vision, sensor fusion and robotics hardware integration, including actuators, cameras and sensors.
  • Knowledge of motion planning and control techniques such as SLAM, path planning, PID control and reinforcement learning. Familiarity with simulation environments, embedded systems and edge AI deployments.
  • Problem-solving and analytical skills, and the ability to collaborate across hardware, AI and software engineering teams.

Preferred

Experience with agentic AI frameworks such as LangGraph, CrewAI and MCP is a plus. The posting offers robotics, AI and edge-computing training, performance-based incentives, health insurance and wellness support, project-based allowances, and a cab facility for women employees.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonC++ROSComputer visionOpenCVPyTorchTensorFlowSensor fusionLiDARIMUGPSMotion planningSLAMPath planningPID controlReinforcement learningRobotics hardware integrationActuatorsCamerasEmbedded systemsEdge AIGazeboWebotsIsaac SimUnity for RoboticsVision AIObject detectionObject trackingAnomaly detectionLarge language models (LLMs)

Your skill match

Checking your profile…

YOUR NEXT STEP

Get interview-ready for this role

A focused preparation guide, built around this job’s responsibilities and requirements.

✦ AI-generated guide
Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.
01

Understand the role

What matters most

  • Build integrated robotic systems

    The role designs and programs humanoid and autonomous platforms combining mechanical components, sensors, cameras, actuators, embedded controllers, and Vision AI. Prepare to explain how you would bring those parts together from prototype toward production.

  • Develop navigation and manipulation control

    The posting calls for algorithms for navigation, motion planning, and object manipulation, with knowledge of SLAM, path planning, and PID control. Prepare to explain design choices and how you would assess control performance.

  • Connect AI perception to robot actions

    The role connects detection, tracking, and anomaly-detection models to control loops and uses agentic workflows and LLMs for multimodal reasoning. Prepare to distinguish what needs a real-time control response from what can use higher-level AI reasoning.

  • Validate safe operation beyond simulation

    The engineer prototypes in simulation, tests in real-world conditions, and supports reliable, safe, compliant deployments. Prepare to describe how test results would inform changes before field use.

  • Optimize and improve field deployments

    The posting specifies edge hardware optimization, collaboration with Platform and AgentOps teams, and feedback loops for improvement. Prepare to discuss resource-aware deployment and how field observations could guide iteration.

Likely challenges to think through

  • Closing the simulation-to-field gap

    A likely difficulty is that a controller that works in simulation may behave differently with physical sensors, actuators, and changing surroundings. Prepare an example validation plan that compares simulated and real-world performance and identifies when to revise the design.

  • Balancing autonomy with predictable control

    Linking Vision AI and LLM-based reasoning to robotic actions may create a trade-off between flexible decisions and timely, reliable responses; this is a preparation hypothesis, not a reported problem. Prepare to explain how you would bound AI-driven actions and test behavior relevant to safety.

  • Fitting capabilities onto edge hardware

    Running perception and control on devices such as Jetson, Raspberry Pi, FPGAs, or custom boards may require trading model complexity against latency and hardware resources. Prepare to explain what you would measure before choosing an optimization.

03

Practice your answers

Technical practice

How would you structure a ROS-based system that connects perception, planning and control?

What to coverThe role calls for Python, C++ and ROS across these functions. Suggested preparation: explain how you would separate components, pass sensor data and commands between them, and decide where timing requirements favor C++ over Python.

How would you integrate cameras, sensors, actuators and an embedded controller on a robot?

What to coverThe posting explicitly calls for robotics hardware integration. Suggested preparation: describe interface checks, calibration and a stepwise test plan for finding faults between hardware and software.

How would you combine LiDAR, IMU and GPS data for localization?

What to coverThe requirements name these sensors, sensor fusion and SLAM. Suggested preparation: discuss what each sensor contributes, how you would handle inconsistent readings, and how you would assess localization accuracy.

How would you plan and update a route for an autonomous robot?

What to coverNavigation, path planning and motion planning are stated duties. Suggested preparation: distinguish route selection from responding to new obstacles, and explain how you would evaluate whether the robot reached its goal safely.

How would you control a robot's motion during object manipulation?

What to coverThe role includes manipulation and names PID control and reinforcement learning among relevant techniques. Suggested preparation: explain feedback control, tuning and when a learned approach might or might not be justified.

How would you connect object detection and tracking to a robotic control loop?

What to coverThe posting specifically calls for linking Vision AI models, including detection, tracking and anomaly detection, to control. Suggested preparation: address uncertain predictions, processing delays and what the robot should do when visual evidence is unreliable.

Where would an agentic AI workflow or LLM fit in an autonomous robot?

What to coverThe role describes agentic decision-making and LLM-based multimodal perception and reasoning; named agentic frameworks are a plus, not a requirement. Suggested preparation: distinguish high-level reasoning from time-sensitive control and explain how decisions would be constrained and checked before physical action.

How would you use simulation before testing a robot in real-world conditions?

What to coverThe posting names ROS, Gazebo, Webots and Isaac Sim for prototyping and testing, followed by real-world validation. Suggested preparation: describe test scenarios, performance measures and how you would investigate differences between simulated and physical behavior.

How would you optimize robotics software for edge hardware?

What to coverThe deployment duties name Jetson, Raspberry Pi, FPGAs and custom boards. Suggested preparation: discuss measuring compute load, latency and memory use, then weigh performance changes against perception or control quality.

How would you validate and improve a robot after field deployment?

What to coverThe role calls for performance testing, reliability, safety, compliance and feedback loops for continuous improvement. Suggested preparation: explain what evidence you would collect, how you would prioritize failures and how you would verify a change before deploying it.

Behavioural practice

Tell me about a time you coordinated across hardware, AI and software teams.

What to coverThe posting emphasizes cross-functional work, including collaboration with Platform and AgentOps teams on deployment. Use a real situation, your specific actions to resolve differing constraints, and the resulting outcome.

Describe a difficult robotics or integration problem you diagnosed.

What to coverThe role calls for analytical problem-solving across sensors, software and control. Give a real example showing the symptoms, the tests you performed, what you changed and the measured or observed result.

Tell me about a time you took ownership of a system from prototype toward deployment.

What to coverThe advertised work spans concept, testing and production. Use a real example to explain your responsibility, how you handled handoffs or unresolved issues, and what happened.

Describe a time safety or reliability concerns changed a technical decision.

What to coverThe posting explicitly emphasizes reliable, safe and compliant robotics deployments. Explain a real situation, the risk you identified, your decision and the outcome without assuming a particular regulatory standard.

Tell me about a time real-world feedback caused you to revise a technical approach.

What to coverThe role includes field validation and feedback loops for improvement. Give a real example of the initial assumption, the evidence that challenged it, the action you took and the result.

02

Sharpen your knowledge

Topics to prepare

  • Localization, planning and control

    Suggested exercise: Explain how SLAM, path planning and PID control would work together to navigate around an obstacle, including where reinforcement learning might fit. The posting explicitly names these techniques and calls for navigation and motion-planning algorithms.

  • Sensor fusion and perception

    Suggested exercise: Draw a data flow from LiDAR, IMU, GPS and camera inputs to a robot’s state estimate and object detections; identify how uncertain readings could affect decisions. The posting requires sensor fusion and computer vision.

  • Vision-to-action control loops

    Suggested exercise: Trace how object detection, tracking or anomaly detection could trigger and update a manipulation or navigation action. The role explicitly connects Vision AI models to robotic control loops.

  • Agentic and multimodal decision-making

    Suggested exercise: Outline which decisions an agentic workflow or LLM could propose, and which should remain subject to deterministic control or safety checks. The posting calls for agentic workflows and LLM-based multimodal reasoning; the proposed boundary is a preparation suggestion, not a stated design requirement.

  • Simulation, validation and deployment reliability

    Suggested exercise: Define test cases and performance measures for moving a navigation behavior from simulation to real-world conditions, including failure and safety checks. The posting calls for simulation, field validation, and reliable, safe deployments.

Skills to revise

  • Python, C++ and ROS implementation

    Suggested exercise: Build a small ROS publisher and subscriber in Python or C++, then explain how you would test their message flow. The posting explicitly requires strong proficiency in all three.

  • Computer-vision pipeline

    Suggested exercise: Use OpenCV and a vision model to process a short camera sequence, inspect detections or tracks, and describe how their outputs would reach a controller. OpenCV, vision models, cameras and control-loop integration appear in the posting.

  • Robotics hardware integration

    Suggested exercise: Diagram and, if hardware is available, bench-test the path from a sensor or camera through an embedded controller to an actuator; check timing and failure behavior. Hands-on integration of actuators, cameras and sensors is explicitly required.

  • Simulation-based prototyping

    Suggested exercise: Prototype a simple navigation or manipulation scenario in Gazebo, Webots or Isaac Sim and record where the behavior fails. These are named simulation platforms, and the role involves prototyping and testing.

  • Edge deployment and optimization

    Suggested exercise: Profile a robotics or vision workload on an available Jetson or Raspberry Pi, then explain a latency or resource trade-off. The posting calls for software optimization on edge hardware, including those devices.

  • Agentic workflow integration

    Suggested exercise: Sketch a workflow that takes perception output, proposes an action and passes it to a robotic control interface; identify how feedback would update the next decision. The role calls for agentic AI integration and feedback loops. LangGraph, CrewAI and MCP are listed only as pluses, not requirements.

04

Your preparation checklist

  • Prepare a robotics project walkthrough

    If you have a relevant project, outline your contribution from design through testing, including hardware, Python/C++ and ROS integration, results and unresolved limitations. The posting emphasizes robotic systems built from concept toward production.

  • Draft a navigation and control design

    As an interview exercise, sketch how a robot would combine sensors, localization, path planning and control. Explain where SLAM or PID control fits and how you would test failure cases; these techniques appear in the requirements.

  • Map a vision-to-action pipeline

    Prepare a diagram connecting camera input, object detection or tracking, decision-making and a robotic control loop. Mark latency, uncertainty and safety checks. The role explicitly connects Vision AI and agentic decisions to robot behavior.

  • Show simulation and field-validation evidence

    Prepare test results or a proposed test plan using a stated platform such as Gazebo, Webots or Isaac Sim. Include measures, real-world validation steps and reliability or safety checks; the posting calls for simulation and field testing.

  • Explain an edge deployment trade-off

    Outline how you would profile and adapt robotics software for a listed target such as Jetson or Raspberry Pi, including compute limits and a feedback loop after deployment. Edge optimization and collaboration with Platform and AgentOps are stated responsibilities.

  • Bring focused questions for the employer

    Ask which robot platforms and edge targets the team uses, how it validates safety and compliance, and how robotics work is divided with Platform and AgentOps. Clarify the experience range: Job Information says 1–3 years, while the opportunity text says 2–5 years.

AI-generated from the job information · Prepared Sep 30, 2026 (UTC)Useful preparation, not a guarantee of eligibility or an offer.

Availability last checked: 4 Nov 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs
Finding similar opportunities…