Full-time · Remote (India)

AI Supply Chain & Data Ops Engineer

Cehpoint is hiring for this role: full-time, Remote (India), ₹12,00,000 – ₹20,00,000 / year. Build and scale the data pipelines and MLOps infrastructure that power Cehpoint’s AI security products.

About the role

You will own the data supply chain that feeds Cehpoint’s AI-driven threat detection and vulnerability-assessment systems — from ingestion and governance through model training and production serving. Your work ensures data reliability, reproducibility, and low-latency delivery so every model iteration is backed by trusted, auditable pipelines.

What you will do

  • Design and maintain ETL/ELT pipelines for ingesting, validating, and transforming threat intelligence feeds, container images, and network telemetry into feature stores.
  • Orchestrate data workflows using Apache Airflow or Prefect, implementing SLA monitoring, retries, dead-letter handling, and alerting on pipeline failures.
  • Containerize training and inference workloads with Docker, package them on Kubernetes, and manage resource allocation across GPU/CPU nodes.
  • Version datasets, schemas, and model artifacts via DVC or MLflow; ensure lineage and auditability for compliance (ISO 27001, SOC 2, DPDP Act 2023).
  • Implement CI/CD for data and ML pipelines, including automated tests, linting, schema validation, and rollback strategies.
  • Collaborate with the security engineering team to harden pipeline environments — secrets management, least-privilege IAM, network segmentation, and encrypted data-at-rest/in-transit.

What you should know

  • Relational and NoSQL data modeling, normalization/denormalization trade-offs, and query optimization fundamentals.
  • MLOps lifecycle: data versioning, feature store concepts, model registry, drift detection, and continuous training patterns.
  • Cloud-native architecture: managed compute, object storage, message queues, and serverless functions on AWS or GCP.

What you should be able to do

  • Python (pandas, PySpark or similar) and advanced SQL
  • Docker & Kubernetes (kubectl, Helm basics)
  • Apache Airflow or Prefect for workflow orchestration
  • MLflow or DVC for experiment & artifact tracking
  • AWS S3 / BigQuery or GCP Cloud Storage / BigQuery for data storage and querying

How you work

  • Hunts for silent data-quality failures — you instinctively add assertions and observability before they become incidents.
  • Reads release notes and RFCs for every tool in the stack; you prototype new features before they hit production.
  • Questions the 'default' pipeline design and pushes for cheaper, faster, or more resilient alternatives.

Skills we check

  • Python & SQL
  • Docker & Kubernetes
  • Apache Airflow or Prefect
  • MLflow / Weights & Biases
  • AWS / GCP cloud services
  • Git & CI/CD (GitHub Actions / GitLab CI)
  • Data warehousing (BigQuery / Snowflake)
  • Linux system administration

What you get

  • Competitive annual stipend (₹12L – ₹20L) with performance-based acceleration.
  • Remote-first setup — your workstation and home-office allowance covered.
  • Generous GPU cloud credits for experimentation and personal projects.
  • Direct mentorship from senior security engineers and ML architects.
  • Path to full-time offer with equity considerations for top performers.

How we select for this role

We read every application ourselves. Here is what happens after you apply.

  1. Apply with a short form

    Paste a link to your CV and answer five questions in your own words. No sign-in and no file upload. About fifteen minutes.

  2. Integrity check

    When you press Check my answers, an automated check confirms the answers were typed by you. You see the result straight away, before you submit.

  3. Choose your project date

    If no cheating is detected you are shortlisted, and we email you a link to choose the day your project starts.

  4. 5-day practical project

    On that day our team contacts you with the brief: real work, 5 working days, reviewed by an engineer. For this role: Over a 5-day milestone, you will build an end-to-end MLOps pipeline: ingest a synthetic threat-intelligence dataset (CSV + JSON), validate and transform it with Python/SQL, orchestrate the workflow with Airflow, track experiments and artifacts with MLflow, containerize the training script with Docker, and deploy a minimal inference endpoint on Kubernetes. The pipeline must include error handling, schema validation, a README with runbooks, and a short demo recording.

  5. Offer

    If your work solves the problem well, you receive your offer letter by email and in the portal.