Question library
Explained questions for the decisions developers make.
Gateways, routing, fallback, quotas, and observability
An AI gateway sits between your application and model providers. It can normalize access, choose approved models, enforce quotas, collect usage, and handle eligible fallback. It is similar to an API gateway specialized for model traffic.
LiteLLM, Portkey, and cloud AI gateways
LiteLLM provides model-access abstraction and proxy/gateway capabilities. Portkey provides an AI gateway and operational controls. Cloud gateways can apply access and model traffic policies within an existing cloud API-management environment; exact AI…
Vision and image understanding
Vision models interpret image content, such as objects, layout, handwriting, and visual relationships. Image understanding produces descriptions or structured observations; it does not automatically verify authenticity or determine business liability. Quality…
PDFs, OCR, and document AI
Document AI combines text extraction, OCR, layout/table understanding, and field extraction. A PDF may already contain usable text or require image interpretation. OCR reads characters; document understanding connects them to fields and structure.
Speech-to-text, text-to-speech, and audio understanding
Speech-to-text converts spoken words into text. Text-to-speech produces spoken output from text. Audio understanding can also process nontext information in audio, depending on the model, but reliable interpretation of emotions or identity should not be…
Voice agents and video understanding
A voice agent handles a live conversation, including listening, responding, interruption, and tools. Video understanding combines information from sampled frames, timestamps, and sometimes audio. Both involve timing and event coordination beyond an ordinary…
Local LLMs, API models, and self-hosting
Local or self-hosted models run on infrastructure you control instead of a provider's shared inference API. An API model outsources much of inference operation. Self-hosting can mean a laptop, on-premises servers, or controlled cloud GPU infrastructure.
Ollama, Hugging Face, and Transformers
Ollama simplifies running supported models locally through a runtime and API. Hugging Face hosts models, datasets, and libraries; Transformers is its widely used library for loading and using supported model architectures. A model repository is a distribution…
vLLM and GPU serving
vLLM is an inference-serving system designed for efficient model execution. Serving techniques such as batching and attention-memory management help multiple requests share expensive hardware. GPU optimization is matching workload, memory, scheduling, and…
Quantization, GGUF, and local embeddings
Quantization represents model values with fewer bits to reduce memory and sometimes improve speed. GGUF is a model file format used by compatible local inference runtimes; it is not itself a model or a guarantee of quantization level. Local embeddings run an…
Fine-tuning and when it is justified
Fine-tuning trains an existing model further using task-specific examples. It changes learned behavior rather than simply adding a document to the current prompt. It can target formatting, domain task performance, style, or consistent instruction following.
LoRA, QLoRA, and PEFT
PEFT means parameter-efficient fine-tuning: update a smaller subset of parameters instead of the whole model. LoRA trains small low-rank adapter matrices attached to selected model operations. QLoRA combines adapter training with a quantized frozen base model…
Dataset preparation, evaluation, and RAG comparison
Dataset preparation converts representative examples into the selected training format with reviewed expected outputs. Evaluation compares trained behavior against the baseline on separate data. RAG supplies current external knowledge at runtime; fine-tuning…
Azure OpenAI and Azure AI Foundry / Microsoft Foundry
Azure OpenAI provides supported OpenAI model access through Azure's service environment. The platform historically called Azure AI Foundry is documented under Microsoft Foundry in the checked official pages. It groups model and AI application capabilities; it…
Azure AI Search, Functions, Container Apps, Key Vault, and Entra ID
Azure AI Search provides searchable indexes and retrieval capabilities. Functions runs event-driven code; Container Apps runs containerized services/jobs. Key Vault manages secrets/keys, while Entra ID provides identity capabilities and access integration.
AWS Bedrock, Lambda, ECS, S3, OpenSearch, and IAM
Bedrock provides managed access to supported foundation models and related capabilities. Lambda runs event-driven functions; ECS orchestrates containers; S3 stores objects such as original PDFs. OpenSearch provides search infrastructure, while IAM controls…
Google Cloud: Vertex AI, Cloud Run, and Cloud Functions
Vertex AI is the familiar Google Cloud AI platform name used in many existing projects; its model documentation currently routes into evolving Google enterprise agent-platform documentation. Cloud Run executes containerized applications and workloads. Cloud…
Shared MERN architecture
flowchart TD
Difficulty is an editorial estimate. Completion means you chose to mark a question as understood.