Technology – GK Questions

Model inversion exploits model confidence scores to infer sensitive attributes of training data. Mitigation: differential privacy, output perturbation, access controls.

Adversarial defenses: adversarial training includes perturbed examples, defensive distillation smooths outputs, randomized smoothing adds noise. All improve robustness against adversarial attacks.

Container image scanning identifies known vulnerabilities, misconfigurations, and secrets in images before deployment. Critical for secure CI/CD pipelines in cloud-native environments.

Service mesh (Istio, Linkerd) automatically encrypts service-to-service traffic with mTLS, provides authentication, authorization, and observability without application code changes.

Kubernetes security: Pod Security Policies restrict privileges, Network Policies control traffic, Resource Limits prevent DoS. Combined for defense-in-depth in container environments.

Device posture assessment evaluates endpoint security: OS patches, antivirus status, encryption, compliance. Critical for Zero Trust access decisions in hybrid environments.

Micro-segmentation creates granular security policies around individual workloads/applications, containing breaches and preventing attackers from moving laterally across the network.

Zero Trust combines: never trust always verify (continuous authentication), assume breach (detect/contain threats), least privilege (minimum access). All principles work together for comprehensive security.

Chain-of-Thought prompting generates reasoning steps.

Fine-tuning updates weights on task-specific data.

Transformer uses self-attention for parallel sequence processing.

Clustering discovers natural groupings in unlabeled data.

Feature engineering: transformations, interactions, encoding.

Time series forecasting models temporal patterns.

Serverless query engines query object storage directly.

Data lakes store raw data; warehouses store curated data.

Parquet is columnar storage optimized for analytics.

Stream processing handles continuous data flows.

CDC streams database operations for real-time analytics.

Kafka, Flink, Spark Streaming enable real-time analytics.