Skip to content
tezvyn:

🤖AI & ML

Artificial intelligence, machine learning, and data science

615 bites

Test yourself: Top 30 intermediate AI & ML interview questionsMultiple choice, with the correct answer and why it is correct on every question. Free, no sign-in.

Intermediate everything in AI & ML

intermediate2 min read

Epsilon in differential privacy and its trade-offs

Epsilon is the privacy budget bounding how much one record can change outputs; smaller epsilon means stronger privacy but more noise and lower accuracy.

intermediate2 min read

Declarative vs imperative ML platform design

Declarative GitOps gives auditable, reproducible, reviewable desired-state config with strong governance but a steeper learning curve; imperative SDKs are flexible and fast for scientists but harder to…

intermediate1 min read

Flask/Gunicorn vs Triton/TorchServe for serving

Flask is simple and flexible but lacks dynamic batching, GPU scheduling, and multi-model management; Triton/TorchServe add those plus metrics and versioning.

intermediate2 min read

Hybrid parallelism for large-model training

Split the model itself via tensor or pipeline parallelism so each replica is smaller, shrinking gradient all-reduce; combine with data parallelism in 2D/3D.

intermediate1 min read

Why design ML pipeline steps to be idempotent?

Re-running a step with the same input yields the same result and no duplicate side effects; enables safe retries and backfills.

intermediate1 min read

Concept drift vs data drift in production models

Data drift is a shift in input distribution P(X); concept drift is a shift in the relationship P(Y|X).

intermediate1 min read

What is a model registry and how does it enable CD?

A registry versions models with metadata, lineage, and stage tags; CD watches stage transitions to trigger deploys.

intermediate1 min read

What is shadow deployment for ML models?

New model receives mirrored live traffic but its predictions are logged, not served; validates real-world behavior and latency without user risk.

intermediate1 min read

Three ways to cut LLM inference cost

Quantization to shrink the model, continuous batching with paged attention to raise GPU utilization, and distillation or smaller routed models for easy queries.

intermediate1 min read

Serving for online and batch predictions

A low-latency online path (synchronous, autoscaled, real-time features) and a high-throughput batch path (parallel, cost-optimized, large jobs) sharing one model artifact and feature definitions.

intermediate1 min read

Diagnosing poor distributed training scaling

Communication overhead (gradient all-reduce, interconnect), data-loading starvation, load imbalance, and small per-GPU batches; profile with the PyTorch profiler and Nsight.

intermediate1 min read

Sub-20ms online feature serving

An in-memory key-value store (Redis) as the online feature store, precomputed features, streaming updates, and offline-online consistency.

intermediate1 min read

Versioning a 10TB dataset as code

Treat data like code via content-addressed pointers in Git while bytes live in object storage; dedupe by hashing so versions share unchanged files.

intermediate1 min read

Directed Acyclic Graph (DAG) for Workflows

A DAG models a workflow as tasks (nodes) connected by dependency edges with no cycles, so a scheduler knows the valid execution order. It enables parallelism, safe retries, and idempotent reruns, and underpins orchestrators like Airflow for ML pipelines.

intermediate1 min read

Staging Environments for ML Pipelines

A staging environment mirrors production so models and pipelines are validated on production-like data and infrastructure before release. It catches drift, integration breaks, and serving regressions early, making promotion to production a safe, repeatable…

intermediate1 min read

AWS Bedrock versus a direct provider API

Bedrock unifies many models with IAM, VPC, and cloud integration; a direct provider API gives earliest models, full feature parity, and simpler vendor terms.

intermediate1 min read

Designing an LLM red-teaming framework

Taxonomy of harms, automated adversarial prompt generation via attacker models and mutation, a classifier to triage outputs, and severity-by-likelihood prioritization.

intermediate1 min read

Scalable oversight of superhuman models

Humans cannot judge outputs beyond their expertise, so feedback degrades; techniques like AI debate or recursive reward modeling decompose judgment.

intermediate1 min read

The alignment tax and capability trade-offs

Alignment tax is capability lost from safety tuning, measured as benchmark or task-success deltas before and after; a product decision weighs over-refusal against harm risk.

intermediate1 min read

Penalizing sycophancy in a reward model

Sycophancy is reward proxy gaming where agreeableness substitutes for correctness; counter it with truth-anchored labels, perturbed-premise pairs, and consistency checks.

We are hiring for this. Every open role lists the topics its interview covers, so you can prepare for the real thing rather than guessing.

See open roles