More in AI & ML — page 30
Describe a grayscale histogram and its use in exposure and equalization
Tests pixel distribution intuition. A strong answer covers intensity bin counts, left or right clustering for exposure errors, and CDF-based redistribution for equalization. Red flag: calling equalization min-max stretching without cumulative mapping.
What is the difference between lossy and lossless image compression?
This tests irreversible discard versus perfect reconstruction. A strong answer defines lossy as dropping detail, lossless as fully reversible, names JPEG, PNG, and chooses lossless for masters, lossy for web. Red flag: claiming lossless is always smaller.
COCO: The Messy Real-World Vision Benchmark
COCO is the standard benchmark for detecting overlapping objects in cluttered scenes. Use it to test object detectors and segmentation. Strong scores here do not mean your model works on specialized domains like medical or satellite imagery.
Mask R-CNN: Region-Based Detection
Mask R-CNN belongs to the R-CNN family core: selective search over CNN feature maps yields bounding boxes with object categories. Reusing convolutional features for localization, not just classification, is the win.
U-Net: Segmentation with Less Data
U-Net retrofits fully convolutional networks to segment images precisely with fewer training examples. It runs a 512 by 512 frame in under a second on a 2015 GPU, fitting latency-sensitive pipelines.
SSD: Real-Time Detection Without Region Proposals
SSD scores default boxes across multiple scales in one forward pass. It runs real-time robotics and mobile vision where two-stage detectors lag. The footgun is ignoring shallow feature maps, which destroys small object accuracy as early layers carry fine…
FLANN Matcher for Feature Correspondence
OpenCV's FLANN matcher pairs query and train descriptors to find cross-image feature correspondences as an alternative to Brute-Force. Engineers often assume FLANN shares Brute-Force's normType and crossCheck parameters, causing silent configuration errors…
MLOps: When to Build vs. Buy Your Infrastructure
Deciding to build or buy MLOps tools hinges on whether it creates a competitive advantage. For commodity tasks like experiment tracking, buying a managed service avoids locking up engineers.
TensorFlow Serving: A Production Server for ML Models
Think of TensorFlow Serving as a dedicated web server for your ML models. It provides a stable API for inference and manages model versions, abstracting away deployment complexity. The main footgun is thinking it only serves models; it serves any 'Servable'.
Weights & Biases: MLOps for Experiment Tracking & Evaluation
Weights & Biases is a platform for MLOps, providing experiment tracking, evaluation, and observability for AI models. It helps you develop models and ship LLM applications. The main risk it addresses is losing track of which model version used which data.
Idempotent Data Pipelines: Reruns Without Side Effects
An idempotent pipeline gives the same output for the same input, no matter how many times you run it. This lets you safely retry failed jobs without side effects, which is crucial for scheduled batch inference or feature engineering tasks.
Slash Your Cloud Bill by Taming Data Egress Costs
Data egress is the tax for moving data out of a cloud provider's network, a common cost in MLOps when moving models or datasets. To save money, keep compute and data in the same region. The footgun is forgetting that traffic between regions also counts.
Cloud Cost Anomaly Detection: Finding Waste
Think of it as a smoke detector for your cloud bill, distinguishing 'good' growth-related costs from 'bad' waste. It automatically flags unexpected spending spikes from misconfigurations or bugs.
Reserved Instances vs. Savings Plans: Pre-pay for Cloud Discounts
Think of Reserved Instances and Savings Plans as buying cloud compute in bulk for a discount. You commit to a certain usage level for 1-3 years to save money on steady-state workloads like production databases.
Data Storage Tiering: Pay Only for the Access You Need
Treat data like items in a house: hot, frequently used data on the counter; cool, less-used data in the pantry. Cloud providers use this to price storage, letting you move old logs to cheaper tiers.
Stop Paying for Idle Cloud Resources
Stop paying for idle cloud servers. Automated shutdown is like turning off the lights in an empty office, running compute only when needed. It's ideal for dev environments or scheduled batch jobs. The footgun is applying this to stateful production services.
GPU Utilization: Are You Wasting Your Most Expensive Resource?
GPU utilization isn't just a percentage; it's a measure of your return on investment. It tells you if your expensive hardware is computing or just waiting for data. Use it to diagnose slow training jobs and right-size cloud instances for ML workloads.
Unit Economics: Tying ML Costs to Business Value
Unit economics connect your ML spending to business outcomes. Instead of a total cloud bill, you see cost per prediction or per token. This helps product owners make pricing tradeoffs and engineers spot efficiency gains.
Showback vs. Chargeback: Who Pays for Compute?
Showback tells teams what their resource usage costs; Chargeback makes them pay for it. It's the difference between a receipt and a bill. These models help manage cloud costs, but implementing chargeback without granular tracking leads to disputes.
Cloud Pricing: On-Demand, Reserved, and Spot Instances
Cloud pricing is like booking a flight. Pay full price for flexibility (On-Demand), get a discount for committing (Reserved), or bid on empty seats for a huge discount but risk getting bumped (Spot). This choice dictates your infrastructure cost.