About Kloudspot

Kloudspot builds AI-powered spatial intelligence software. Its LISA platform combines camera, Wi-Fi, BLE and IoT signals to help airports, factories and enterprises understand physical spaces in real time.

About the Role

We are looking for a technically robust Machine Learning Engineer who can work across the complete machine learning lifecycle — from understanding an ambiguous business problem and working with real-world data to developing, deploying and monitoring models in production.

This is a hands-on, Python-heavy role involving sensor data, streaming systems and end-to-end model ownership.

Key Responsibilities

- Translate business problems into well-defined machine learning problems.

- Build and evaluate models for classification, forecasting, clustering, anomaly detection and time-series problems.

- Clean, transform and validate noisy real-world sensor and video data.

- Perform feature engineering and data analysis.

- Build visualisations to understand data quality, model behaviour and failure modes.

- Deploy, monitor and retrain machine learning models in production.

- Make informed decisions around algorithm selection and model performance.

- Debug and troubleshoot ML solutions in production environments.

- Work with streaming data pipelines and production systems

RequirementsRequirements

- 3–5 years of experience in shipping machine learning models to production.

- Strong programming skills in Python.

- Hands-on experience with Num Py, pandas and scikit-learn.

- Strong experience with PyTorch or Tensor Flow.

- Strong understanding of probability, statistics, linear algebra and optimisation.

- Ability to understand why an algorithm works, evaluate different approaches and explain their limitations.

- Strong skills in data cleaning, feature engineering and data visualisation.

- Hands-on experience with MongoDB,



including aggregation pipelines, indexing and querying large collections.

- Working experience with Kafka and streaming data pipelines.

- Practical MLOps experience with Docker, CI/CD, experiment tracking and model monitoring.

- Strong problem-solving and debugging skills with the ability to work independently.

Preferred / Domain Experience

Experience in the manufacturing domain will be an advantage, particularly exposure to:

- Production lines and shop-floor analytics

- OEE

- Yield, throughput or output prediction

- Predictive maintenance

- Equipment alarms

- SPC

- MES, SCADA, PLC or historian data

Nice to Have

- Agentic AI: Agentic flows and multi-agent systems

- LLM / RAG: LLM workflows and RAG pipelines

- Computer Vision: Practical computer vision experience

- Cloud ML: AWS, GCP or Azure ML

- Kubernetes: Container orchestration and production operations

- Edge AI: Model deployment to edge environments

What Success Looks Like

You will be successful in this role if you can confidently take a messy real-world problem, convert it into a defensible ML solution, and take that solution through to a reliable production deployment, while clearly explaining the mathematics, assumptions, trade-offs and failure modes behind your decisions.

BenefitsWhat You Can Expect

- Opportunity to work on real-world machine learning problems involving sensor, video and streaming data.

- End-to-end ownership of machine learning models, from problem definition through production deployment.

- Hands-on exposure to MLOps, streaming systems and production ML environments.

- Opportunity to work with technologies including Python, PyTorch/Tensor Flow, MongoDB, Kafka and Docker.

- Exposure to AI-powered spatial intelligence applications across enterprise environments.

This positioning is supported by the source's description of the role and working environment.

📌 Machine Learning Engineer / Software Engineer (India)
🏢 Kloudspot
📍 India