Get AI summaries of any video or article — Sign up free

The Full Stack — Channel Summaries

AI-powered summaries of 103 videos about The Full Stack.

103 summaries

No matches found.

LLMOps (LLM Bootcamp)

The Full Stack · 3 min read

LLMOps is less about picking the “best” language model and more about building a reliable production loop: choose a model with the right trade-offs,...

LLMOpsModel SelectionPrompt Management

Launch an LLM App in One Hour (LLM Bootcamp)

The Full Stack · 3 min read

Large language models are turning into general-purpose “next-word” engines that can power far more than chat—especially when paired with language...

Language ModelingLanguage User InterfacesRetrieval-Augmented Generation

LLM Foundations (LLM Bootcamp)

The Full Stack · 3 min read

Large language models work because they turn text into numbers, then learn—via gradient-based training—to predict the next token using a Transformer...

Transformer FoundationsAttention MechanismTokenization

Harrison Chase - Agents Masterclass from LangChain Founder (LLM Bootcamp)

The Full Stack · 3 min read

Agent systems are built around a simple but consequential shift: use a language model as a reasoning engine that decides which tool to call next,...

Agent Tool UseReAct PromptingProduction Reliability

Lecture 1: Deep Learning Fundamentals (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Deep learning fundamentals hinge on a simple but powerful idea: neural networks are flexible function approximators whose weights can be trained by...

PerceptronUniversal ApproximationLoss Functions

Lecture 01: When to Use ML and Course Vision (FSDL 2022)

The Full Stack · 3 min read

Machine learning is moving into the mainstream, but the real challenge isn’t getting models to work—it’s deciding when ML is worth the added...

ML-Powered ProductsWhen To Use MLML Ops vs ML Products

Augmented Language Models (LLM Bootcamp)

The Full Stack · 3 min read

Augmented language models hinge on a simple constraint: modern LLMs are strong at language and instruction-following, but they lack up-to-date world...

Augmented Language ModelsRetrieval AugmentationEmbeddings

Chip Huyen on Machine Learning Interviews (Full Stack Deep Learning - November 2019)

The Full Stack · 3 min read

Machine learning hiring is less about “perfect” interviews and more about navigating a noisy, expensive, and often inconsistent process—so candidates...

Machine Learning InterviewsApplied Research vs ResearchResearch Engineer

Learn to Spell: Prompt Engineering (LLM Bootcamp)

The Full Stack · 3 min read

Prompt engineering is the practical art of choosing the exact text you feed a language model so it behaves the way you need—often replacing what used...

Prompt EngineeringConditioningInstruction Tuning

Lecture 1: Introduction to Deep Learning - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Deep learning’s breakthrough in 2012 wasn’t just a better model—it replaced hand-crafted image features with learned representations, turning “what...

Deep Learning FoundationsImageNet BreakthroughRepresentation Learning

Lecture 6: Infrastructure & Tooling (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Deep learning progress depends less on model code than on the surrounding “infrastructure and tooling” that turns raw data into continuously...

MLOps InfrastructureData PipelinesDistributed Training

LangChain Demo + Q&A with Harrison Chase

The Full Stack · 3 min read

LangChain’s core value is turning large language models from “text-in, text-out” into usable applications by providing the missing framework:...

LangChain FrameworkChat Over DocumentsRetrieval Augmented QA

UX for Language User Interfaces (LLM Bootcamp)

The Full Stack · 3 min read

Language user interfaces are poised to become the next major step change in computing—replacing menus, forms, and command buttons with text-first...

Language User InterfacesAffordances and SignifiersAI UX Patterns

1. Overview - ML Projects - Full Stack Deep Learning

The Full Stack · 2 min read

Deep learning projects fail far more often than teams expect—one survey cited in the discussion found that 85% of AI projects at large companies...

ML Project LifecycleProject FeasibilityMetrics and Baselines

Lecture 02: Development Infrastructure & Tooling (FSDL 2022)

The Full Stack · 3 min read

Machine learning development runs on a “data flywheel,” but getting from an idea to a reliable system at scale depends on disciplined software...

Development WorkflowReproducible EnvironmentsDeep Learning Frameworks

Lecture 5: ML Projects (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Machine learning projects fail less because models are “bad” and more because teams start with unclear goals, unrealistic feasibility assumptions,...

ML Project LifecycleProject PrioritizationMetrics Optimization

Lecture 2: Setting Up Machine Learning Projects - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Machine learning projects succeed or fail less on model choice than on how well teams plan, collect data, test beyond validation scores, and set...

Machine Learning Project LifecycleProject PlanningData Collection Labeling

Reza Shabani - How Replit Trained Their Own LLMs (LLM Bootcamp)

The Full Stack · 3 min read

Replit’s Ghostwriter code-completion model is built through a tightly engineered pipeline designed to make smaller, cheaper, and more specialized...

Training Custom LLMsCode Data PipelinesTokenizer Training

Lecture 4: Transfer Learning and Transformers (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Transfer learning is the bridge that lets large, pre-trained neural networks work on small, task-specific datasets—first in computer vision, then in...

Transfer LearningWord EmbeddingsELMo and ULMFiT

What's Next? (LLM Bootcamp)

The Full Stack · 3 min read

Multimodal large language models are rapidly turning into general-purpose “brains” for both software and physical machines—especially robotics—by...

Multimodal TransformersVision TransformersGeneral Purpose Robotics

2. Lifecycle - ML Projects - Full Stack Deep Learning

The Full Stack · 3 min read

Machine learning projects follow a repeatable lifecycle—planning, data collection, training/debugging, and staged deployment—but progress often loops...

Machine Learning LifecycleData CollectionPose Estimation

Lab 04: Experiment Management (FSDL 2022)

The Full Stack · 3 min read

Experiment management is the difference between “useful training output” and “lost knowledge.” During model training, metrics like loss and...

Experiment ManagementTensorBoardWeights & Biases

Lecture 11A: Deploying ML Models (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Machine learning models don’t become “production-ready” just because they work in a notebook; they need a deployment path that fits the latency,...

Model DeploymentBatch PredictionModel Service APIs

Lecture 7: Troubleshooting Deep Neural Networks (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Troubleshooting deep neural networks is hard because the same drop in performance can come from many different causes—and many bugs don’t announce...

Neural Network DebuggingBias-Variance DecompositionData Pipeline Bugs

Labs 1-3: Introduction to the Text Recognizer Project - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Handwritten-text recognition is built as a full pipeline: a web backend accepts an encoded image, a deployed “compiled prediction model” runs...

Text Recognizer ArchitectureEMNIST Character TrainingCTC Line Recognition

Lecture 2A: Convolutional Neural Networks (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Convolutional neural networks gained their edge in computer vision by replacing the “flatten an image and learn a giant matrix” approach with a...

Convolutional FiltersStride And PaddingReceptive Field

Lecture 06: Continual Learning (FSDL 2022)

The Full Stack · 3 min read

Continual learning in production is less about “retraining whenever something feels off” and more about running a structured retraining strategy that...

Continual LearningRetraining StrategyMonitoring & Observability

Lab 02: PyTorch Lightning and Convolutional NNs (FSDL 2022)

The Full Stack · 2 min read

PyTorch Lightning is presented as the practical fix for the “sharp edges” of hand-rolling PyTorch training loops—especially when training needs to...

PyTorch LightningConvolutional Neural NetworksTraining Checkpoints

Lecture 3: Recurrent Neural Networks (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Recurrent neural networks (RNNs) were built to handle sequence data efficiently by reusing the same weights across time and carrying information...

Sequence ProblemsRecurrent Neural NetworksLSTM Gates

Jeremy Howard on Platform.ai and Fast.ai (Full Stack Deep Learning - March 2019)

The Full Stack · 3 min read

Jeremy Howard argues that “augmented machine learning”—tight human–computer collaboration—beats fully automated ML pipelines for most practical...

Augmented Machine LearningHuman-in-the-Loop LabelingTransfer Learning Defaults

Lecture 2B: Computer Vision Applications (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Computer vision deep learning has advanced largely by swapping in better image-recognition backbones—then reusing those same building blocks for...

ImageNet ClassificationConvolutional BackbonesObject Detection

Lab 07: Web Deployment (FSDL 2022)

The Full Stack · 3 min read

A practical deployment pipeline turns a trained PyTorch text recognizer into a portable, shareable model service—first by compiling it to...

TorchScript ConversionW&B ArtifactsGradio UI And API

Project Walkthrough: askFSDL (LLM Bootcamp)

The Full Stack · 3 min read

A Discord bot built for askFSDL delivers retrieval-augmented question answering over a curated knowledge base, but the biggest gains come less from...

Retrieval-Augmented Q&AETL and ChunkingLangChain Prompting

Lecture 8: Data Management (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Data management is where most deep learning projects quietly win or fail: getting messy, distributed inputs into a GPU-ready training pipeline—and...

Data FlywheelSemi-Supervised LearningData Storage

Lecture 10: ML Testing & Explainability (Full Stack Deep Learning - Spring 2021)

The Full Stack · 2 min read

Machine-learning systems fail in ways that offline test scores can’t fully predict, so teams need a broader testing mindset: validate not just a...

ML TestingPerformance EnvelopeShadow Testing

Lecture 07: Foundation Models (FSDL 2022)

The Full Stack · 3 min read

Foundation models are driving a shift in AI from task-specific systems toward general-purpose models built by scaling architecture, data, and...

Foundation ModelsTransformersScaling Laws

Panel Discussion: Do I need a PhD to work in ML? (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

A PhD is not a universal requirement for working in machine learning; it depends on what kind of ML work someone wants to do and how employers...

PhD RequirementsApplied Machine LearningFrontier Research

Lecture 05: Deployment (FSDL 2022)

The Full Stack · 3 min read

Model deployment is where machine learning stops being a lab exercise and starts proving it can solve real user problems—often revealing flaws that...

Model DeploymentPrototype ToolsBatch vs Online Serving

Lecture 4: Infrastructure and Tooling - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Deep learning success depends less on model architecture than on building an end-to-end system that can ingest data, train reliably, deploy safely,...

ML LifecycleExperiment ManagementDeep Learning Frameworks

Lecture 03: Troubleshooting & Testing (FSDL 2022)

The Full Stack · 3 min read

Troubleshooting and testing in software is about risk reduction, but testing never becomes a guarantee of correctness—so the practical goal is to...

Software TestingML Smoke TestsExpectation Testing

Lecture 04: Data Management (FSDL 2022)

The Full Stack · 3 min read

Data management is the hidden driver of machine-learning performance: spending far more time on data than on models—especially on dataset quality,...

Data ExplorationStorage ArchitectureSQL And Data Frames

3. Prioritizing - ML Projects - Full Stack Deep Learning

The Full Stack · 3 min read

Picking the right machine learning projects comes down to a simple but disciplined tradeoff: pursue work that delivers high business impact while...

ML Project PrioritizationCheap PredictionSoftware 2.0

Lab 06: Data Annotation (FSDL 2022)

The Full Stack · 3 min read

Data annotation is treated as a make-or-break step in the full machine-learning pipeline: rich, carefully structured labels—often at finer...

Data AnnotationLabel StudioSynthetic Data

Lecture 11B: Monitoring ML Models (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Monitoring deployed machine learning models is about catching silent performance decay—often driven by changes in data, user behavior, or sampling...

Model DriftData DriftDistribution Monitoring

Lecture 13: ML Teams (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Machine learning teams fail or succeed less on model quality alone and more on how organizations staff roles, structure accountability, and manage...

Machine Learning TeamsML Organization StructureML Role Definitions

Overview (1) - Infrastructure and Tooling - Full Stack Deep Learning

The Full Stack · 2 min read

Turnitin’s products sit at the intersection of writing support and academic integrity: Revision Assistant provides detailed, non-grading feedback to...

Academic IntegrityWriting FeedbackGrading Automation

Lab 08: Monitoring (FSDL 2022)

The Full Stack · 3 min read

Model monitoring for a production text recognizer has to go beyond infrastructure health checks and into “behavioral” signals—whether the system’s...

Model MonitoringUser FeedbackGantry Logging

6. Baselines - ML Projects - Full Stack Deep Learning

The Full Stack · 3 min read

Baselines act as a reality check for model performance by setting a lower bound on what a system can achieve. The tighter that lower bound, the more...

Model BaselinesHuman PerformanceDebugging

Lab 9: Web Deployment (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Lab 9 turns a trained paragraph text recognizer into something that can be called over HTTP and packaged for deployment. The core move is speeding up...

TorchScriptFlask APIDocker Deployment

Lecture 7: Machine Learning Teams - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Machine learning teams face a widening talent gap that makes hiring—and building effective teams—far harder than most companies expect. Estimates...

AI Talent GapMachine Learning RolesTeam Structure

Livecoding: Getting Started with LLMs, by Jeremy Howard

The Full Stack · 3 min read

The core takeaway is that strong performance on an LLM multiple-choice science benchmark comes less from clever prompting and more from disciplined...

LLM EDAMAP@3 EvaluationGPT-3.5 Baseline

Lab 1 - Introduction - Full Stack Deep Learning

The Full Stack · 3 min read

The lab setup centers on building a production-minded deep learning pipeline for a text-recognition app—turning an uploaded page image into a clean...

Text Recognition ArchitectureLab SetupEMNIST Dataset

Lab 05: Troubleshooting & Testing (FSDL 2022)

The Full Stack · 3 min read

Testing and performance troubleshooting for deep learning systems hinge on two disciplines: automated quality gates for code and data, and a...

Pre-commit Hookspytest TestingMemorization Tests

Richard Socher on NLP at Salesforce (Full Stack Deep Learning - March 2019)

The Full Stack · 3 min read

Natural language processing is stuck in a cycle of single-task models that improve benchmarks but don’t add up to a general system. Richard Socher’s...

Multitask NLPZero-Shot LearningPointer Networks

Peter Welinder - Fireside Chat with OpenAI VP Product (LLM Bootcamp)

The Full Stack · 3 min read

Peter Welinder traces a career path from early confusion about “artificial intelligence” to product-focused machine learning—and credits a series of...

Career PathComputer VisionDeep Reinforcement Learning

Andrej Karpathy on AI at Tesla (Full Stack Deep Learning - August 2018)

The Full Stack · 3 min read

Deep learning for real-world autonomy is shifting the center of gravity from “clever algorithms” to “programming with data.” Andrej Karpathy,...

Software 2.0Data EngineLabel Imbalance

Software Engineering (2) - Infrastructure and Tooling - Full Stack Deep Learning

The Full Stack · 2 min read

Python has become the default language for full-stack deep learning less because it’s inherently perfect for scientific computing and more because...

Choosing PythonJupyter NotebooksStatic Analysis

Lecture 8: Troubleshooting Deep Neural Networks - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Troubleshooting deep neural networks is hard not because training is mysterious, but because the same drop in performance can come from many...

Deep Learning DebuggingBias-Variance DecompositionSingle-Batch Overfitting

Lab 3: RNNs (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Sequence models for handwritten text recognition take a practical turn in Lab 3: a sliding-window CNN baseline quickly works when characters don’t...

Sliding WindowsCTC LossFully Convolutional CNN

Lecture 9: Testing and Deployment - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Machine learning systems need a different testing and deployment playbook than traditional software because the “running system” depends on both code...

ML Testing StrategyProduction MonitoringCI With Docker

Lecture 08: ML Teams and Project Management (FSDL 2022)

The Full Stack · 3 min read

Machine-learning product teams face a structural problem: ML adds uncertainty, scarce talent, and stakeholder misunderstanding on top of the usual...

ML Team RolesML HiringML Organization Archetypes

4. Archetypes - ML Projects - Full Stack Deep Learning

The Full Stack · 3 min read

Machine learning projects tend to fall into three archetypes—improving an existing process, augmenting a manual workflow, or automating a manual...

Machine Learning ArchetypesData FlywheelDownstream Metrics

Lecture 6: Data Management - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Data management in deep learning is less about model math and more about building a reliable pipeline for labels, storage, versioning, and...

Data FlywheelLabeling InterfacesData Storage

Lukas Biewald on Founding Weights & Biases and FigureEight (Full Stack Deep Learning - March 2019)

The Full Stack · 3 min read

Deep learning’s real bottleneck isn’t model architecture—it’s the messy, high-stakes work of turning training into reliable production systems. Lucas...

ML DeploymentTraining DataModel Generalization

5. Metrics - ML Projects - Full Stack Deep Learning

The Full Stack · 3 min read

Choosing the right metric is the make-or-break decision that determines whether an ML project can be steered toward real-world usefulness. Because...

Metric SelectionPrecision and RecallThresholding Strategies

Computing and GPUs (3) - Infrastructure & Tooling - Full Stack Deep Learning

The Full Stack · 3 min read

Deep learning progress over the past five years has tracked compute growth closely enough that hardware choices now shape what experiments are even...

Compute InfrastructureGPU SelectionMixed Precision

Labs 4-5: Tracking Experiments - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Handwriting line recognition is built from two linked pieces: a convolutional network that scans an input line image window-by-window, and a sequence...

Handwriting Line RecognitionSliding Window CNNCTC and LSTM

Pieter Abbeel on Research Directions (Full Stack Deep Learning - November 2019)

The Full Stack · 3 min read

Research frontiers in deep learning are increasingly about learning systems that can adapt quickly—often with only a few examples or trials—while...

Few-Shot LearningModel-Agnostic Meta-LearningMeta Reinforcement Learning

Lecture 12: Research Directions (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Deep learning research is shifting from “interesting ideas” to “rapidly deployable tools,” and the lecture’s through-line is that the fastest...

Research DirectionsUnsupervised LearningContrastive Learning

Lecture 9: Ethics (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Ethics in machine learning isn’t about “feeling” that something is right or simply following the law. It’s about making defensible choices under...

Ethical FrameworksAI AlignmentFairness Metrics

Labeling (3) - Data Management - Full Stack Deep Learning

The Full Stack · 3 min read

Data labeling hinges less on the annotation software’s feature list and more on the human decisions inside the labeling workflow—especially when...

Data LabelingAnnotation QualityLabeling Labor

Roles (2) - ML Teams - Full Stack Deep Learning

The Full Stack · 3 min read

Machine learning teams split work across distinct roles—ML product management, DevOps, data engineering, ML engineering, ML research, and data...

ML Team RolesML LifecycleData Labeling

Why you should always overfit a single batch to debug your deep learning model

The Full Stack · 2 min read

Debugging a deep learning model becomes dramatically easier once training runs end-to-end and the system can overfit a single batch. The core idea is...

DebuggingOverfittingLoss Behavior

Lab 8: Testing and Continuous Integration (Full Stack Deep Learning - Spring 2021)

The Full Stack · 3 min read

Lab 8 focuses on making a full-stack handwriting OCR project safer to change by adding automated linting, targeted tests, and continuous integration....

Testing StrategyLinting PipelineCharacter Error Rate

What machine learning role is right for you?

The Full Stack · 3 min read

Machine learning teams hire for several distinct roles—DevOps, data engineering, machine learning engineering, machine learning research, and data...

Machine Learning RolesDevOpsData Engineering

ML Test Score (2) - Testing & Deployment - Full Stack Deep Learning

The Full Stack · 2 min read

Machine learning systems accumulate “hidden technical debt” because the work doesn’t end at model training. Once a model is deployed, it becomes a...

ML Test ScoreProduction ML TestingData Skew Monitoring

Frameworks & Distributed Training (5) - Infrastructure & Tooling - Full Stack Deep Learning

The Full Stack · 3 min read

Deep learning frameworks have shifted from “fast in production, painful in development” toward a convergence where developers write in Python with...

Deep Learning FrameworksDistributed TrainingData Parallelism

Managing (4) - ML Teams - Full Stack Deep Learning

The Full Stack · 3 min read

Managing machine-learning teams is hard largely because progress is unpredictable: early gains often don’t translate into sustained improvement, and...

ML Team ManagementProbabilistic PlanningResearch vs Engineering

Project Structure (1) - Testing & Deployment - Full Stack Deep Learning

The Full Stack · 2 min read

A practical full-stack deep learning setup hinges on separating three systems—prediction, training, and serving—and then testing each with the right...

Project StructureTesting StrategyModel Serving

Hyperparameter Tuning (7) - Infrastructure and Tooling - Full Stack Deep Learning

The Full Stack · 2 min read

Hyperparameter tuning is often where deep-learning experiments stall: teams can guess a rough model size, but the real question is how to search the...

Hyperparameter OptimizationHyperoptSigOpt

Overview (1) - ML Teams - Full Stack Deep Learning

The Full Stack · 2 min read

Machine learning teams are unusually difficult to run because every core responsibility of technical management—hiring, alignment of work, long-term...

ML Team ManagementRole LandscapeTechnical Debt

All in One (8) - Infrastructure and Tooling - Full Stack Deep Learning

The Full Stack · 3 min read

The push toward “all-in-one” deep learning infrastructure is about replacing a patchwork of point tools with a single system that can take models...

All-in-One ML PlatformsExperiment TrackingModel Deployment

Docker (4) - Testing & Deployment - Full Stack Deep Learning

The Full Stack · 3 min read

Docker’s core value is that it packages an application with only the binaries and libraries it needs—no guest operating system—making deployments...

Docker vs Virtual MachinesDockerfile LayersDocker Hub Registry

Resource Management (4) - Infrastructure & Tooling - Full Stack Deep Learning

The Full Stack · 2 min read

Resource management in deep learning is about making shared compute usable: multiple people need to launch experiments quickly, with dependencies...

Resource ManagementSLURM SchedulingDocker Containers

Sources (2) - Data Management - Full Stack Deep Learning

The Full Stack · 3 min read

Deep learning in production often hinges less on flashy model design and more on how teams source, label, and multiply data. Label-hungry approaches...

Data FlywheelSemi-Supervised LearningData Augmentation

Versioning (5) - Data Management - Full Stack Deep Learning

The Full Stack · 3 min read

Versioning in machine learning isn’t just about saving model code—it’s about making the trained artifact reproducible by tracking the exact data used...

Data VersioningML ReproducibilityGit LFS

Debug (3) - Troubleshooting - Full Stack Deep Learning

The Full Stack · 3 min read

Debugging deep learning starts with a practical goal: make the model run end-to-end, then prove it can learn by forcing it to overfit a single batch,...

Deep Learning DebuggingTensor ShapesOverfitting Single Batch

Lecture 09: Ethics (FSDL 2022)

The Full Stack · 3 min read

Ethics in tech and machine learning comes down to managing three recurring tensions—alignment failures, stakeholder trade-offs, and the need for...

Ethics in TechnologyMachine Learning FairnessDark Patterns

Storage (4) - Data Management - Full Stack Deep Learning

The Full Stack · 3 min read

Storage choices determine how data moves, how fast it can be read, and how safely it can be reused across training and production. The core takeaway...

File SystemsObject StorageDatabases

Orgs (3) - ML Teams - Full Stack Deep Learning

The Full Stack · 3 min read

Machine-learning organizations don’t have a single “correct” structure yet, but companies tend to evolve through a recognizable ladder: from ad hoc...

Machine Learning TeamsOrganizational StructureCentralized vs Embedded ML

Start Simple (2) - Troubleshooting - Full Stack Deep Learning

The Full Stack · 3 min read

Starting simple is the fastest way to find out whether poor model performance comes from a hard problem or from avoidable bugs in the pipeline. The...

Troubleshooting WorkflowBaseline ArchitecturesHyperparameter Defaults

Deep Learning Frameworks

The Full Stack · 2 min read

Deep learning frameworks can be judged along two practical axes: how pleasant they are for building models and how well they scale once those models...

Framework TradeoffsCaffeTensorFlow

CI/Testing (3) - Testing & Deployment - Full Stack Deep Learning

The Full Stack · 2 min read

Continuous integration is the backbone of reliable machine-learning development: every time code is pushed, an automated pipeline runs tests (and...

Continuous IntegrationUnit vs Integration TestsContainerization

Lecture 10: Research Directions - Full Stack Deep Learning - March 2019

The Full Stack · 3 min read

Research momentum in deep learning has accelerated to the point where thousands of papers arrive every month, making it impossible for any one person...

Few-Shot LearningModel-Agnostic Meta-LearningReinforcement Learning

Monitoring (6) - Testing & Deployment - Full Stack Deep Learning

The Full Stack · 3 min read

Monitoring for machine learning deployments isn’t just about keeping servers alive—it’s about catching data and model failures early, then feeding...

Model MonitoringData DriftBusiness Metrics

Processing (6) - Data Management - Full Stack Deep Learning

The Full Stack · 2 min read

Building a photo popularity predictor that updates daily forces data pipelines to do more than just “run a model.” The core need is reliable data...

Data WorkflowsAirflow DAGMakefile Dependencies

Hiring (5) - ML Teams - Full Stack Deep Learning

The Full Stack · 2 min read

AI hiring is being squeezed by a widening talent gap: estimates suggest only thousands to a few hundred thousand people can build AI systems, far...

AI Talent GapML HiringSourcing Strategies

Evaluate (4) - Troubleshooting - Full Stack Deep Learning

The Full Stack · 3 min read

Model improvement starts with evaluation, not guesswork: once a team is reasonably confident the model is bug-free, the next move is to measure...

Bias–Variance DecompositionDistribution ShiftModel Evaluation

Improve (5) - Troubleshooting - Full Stack Deep Learning

The Full Stack · 3 min read

Model improvement starts with a simple priority order: fix underfitting first, then tackle overfitting, and only after both training and validation...

Bias-Variance PrioritizationUnderfitting RemediesOverfitting Regularization

Tune hyper-parameters (6) - Troubleshooting - Full Stack Deep Learning

The Full Stack · 3 min read

Hyper-parameter tuning is the last major lever after training and validation curves look “reasonably close,” but it’s hard because there are many...

Hyper-parameter TuningLearning Rate SchedulesRandom Search