AEO (answer engine optimisation)
Structuring content to give direct answers that search features and AI assistants can quote, such as clear definitions, lists and FAQs.
Plain definitions from the engineers who use them, with links to where we put each one into practice.
143 terms8 categoriesBrowse by topic
Structuring content to give direct answers that search features and AI assistants can quote, such as clear definitions, lists and FAQs.
AI systems that plan, act through tools and check the result over several steps with limited human input. The term covers single agents and teams of agents working together.
Software that uses a language model to decide which steps and tools to use to reach a goal, instead of following a fixed script.
At Techdome: Aether Voice, our voice agent product, answers and qualifies calls around the clock, with knowledge-graph memory and calendar sync.
A spreadsheet-style database with forms, views and automations, often used for lightweight internal tools.
AWS object storage for files, backups and static assets, with very high durability.
A PostgreSQL extension that adds graph database features, so graph queries can run alongside relational data.
At Techdome: Our GraphRAG systems use Apache AGE on PostgreSQL.
A distributed event-streaming platform for moving large volumes of messages between systems reliably and in order.
Connecting two systems so they exchange data through their APIs, instead of people copying data between them.
Checking that an API returns the right data and errors for different requests, without going through a user interface.
At Techdome: Sparrow, our open-source API testing suite, was Product of the Day #4 on Product Hunt twice.
A web framework for content-focused sites that renders pages to HTML and ships little JavaScript by default, adding interactivity only where needed.
A tamper-resistant record of who did what and when in a system, needed in healthcare, finance and other regulated work.
An identity platform that handles sign-in, single sign-on and user management, so teams do not build authentication themselves.
Amazon's cloud platform for computing, storage, databases and AI services.
At Techdome: Our telehealth platform runs on AWS HIPAA infrastructure.
ECS and EKS run containers (ECS is the AWS-native orchestrator, EKS is managed Kubernetes). Lambda runs small functions on demand without managing servers.
An open-source engine that tests web pages for accessibility problems automatically, usable in a browser or a CI pipeline.
OpenAI models offered through Microsoft Azure, with Azure's enterprise security, networking and compliance controls.
A contract under HIPAA between a healthcare organisation and a vendor that handles patient data for it, setting how that data must be protected.
A Node.js job queue library built on Redis, for running background tasks with retries, delays and rate limits.
Continuous integration and delivery: automated steps that build, test and deploy code on every change, so releases are small and frequent.
A family of large language models from Anthropic.
Chat and voice assistants that hold a natural back-and-forth with a user and keep context across turns.
Google's measures of real-user page experience: loading speed, responsiveness and visual stability.
A monitoring and observability platform for metrics, logs and traces.
A group of engineers who work only on your product over a long period, usually billed monthly.
At Techdome: Our pods are a lead architect, senior full-stack engineers and a QA lead, on monthly retainers.
A speech-to-text service with real-time transcription APIs.
Combining software development and operations so code is built, tested and released automatically and reliably.
At Techdome: Our delivery stack uses Docker, Kubernetes and Terraform on AWS and Azure.
An early stage of research, scoping and design that turns an idea into a clear plan and estimate before building starts.
At Techdome: Week 1 of our Blueprint is de-risking and architecture, before core engineering begins.
A tool that packages an application and its dependencies into a container that runs the same way everywhere.
A framework for building desktop apps with web technologies, packaging a web app with its own browser engine and Node.js.
A text-to-speech service that produces natural-sounding synthetic voices.
Lists of numbers that represent the meaning of text, images or other data, so similar items sit close together and can be found by similarity search.
The European accessibility standard for ICT products and services, used in public procurement, which incorporates WCAG.
At Techdome: We audited a UAE public-sector tourism programme against EN 301 549 and WCAG 2.2 AA.
A small, self-managing team, usually an architect, engineers and QA, that owns delivery of a product or part of one.
At Techdome: A Techdome pod is embedded in your sprint cadence and Slack in about two weeks.
Extract, transform, load: pulling data from sources, cleaning and reshaping it, and loading it into a database or warehouse for reporting.
A minimal, widely used web framework for Node.js for building APIs and web servers.
A modern Python framework for building APIs, using type hints for validation and generating documentation automatically.
A fast, low-overhead Node.js web framework with schema-based validation and a plugin system.
Further training a pre-trained model on your own examples so it follows a style or task more consistently. Better prompting or retrieval is often cheaper and enough first.
An agreement for a defined scope at an agreed price. Predictable for the buyer, but changes to scope are renegotiated.
At Techdome: Our fixed-scope sprints run two to four weeks and start at $8,000.
A senior technology leader who works part-time for a company, setting architecture, hiring and engineering practice without a full-time hire.
A customer support platform that turns emails, chats and calls into tickets.
The EU's data protection law. It needs a lawful basis for using personal data and gives people rights to access and erase it, and it applies to anyone handling EU residents' data.
At Techdome: Our architecture builds in data minimisation and right-to-erasure workflows from day zero.
Google's family of multimodal AI models, which work with text, images, audio and code.
Making a brand's content easy for AI answer engines such as ChatGPT, Perplexity and Gemini to find, understand and cite.
GitHub's built-in automation for building, testing and deploying code whenever changes are pushed.
A marketing and sales platform with a CRM, email, SMS and workflow automation, often used by agencies.
Google's cloud platform for computing, storage, databases and AI services.
A query language for APIs that lets the client ask for exactly the fields it needs in one request.
A form of RAG that also stores the relationships between entities in a knowledge graph, so the system can answer questions that need several linked facts.
At Techdome: Our GraphRAG systems run on PostgreSQL with pgvector and Apache AGE for the knowledge graph.
When a language model states something false with confidence. Grounding answers in sources and checking outputs reduces it, but does not remove it.
The US law that sets rules for protecting patients' health information. Software that handles that data must follow it.
At Techdome: Our telehealth platform was designed for HIPAA and UK GDPR from day zero, with encrypted data, audit logging and tenant isolation.
A CRM and marketing platform for managing contacts, deals and email campaigns.
A property of an operation that gives the same result if it runs more than once, which makes retries and repeated webhook deliveries safe.
A database built for time-series data such as sensor readings and metrics.
Describing servers, networks and services in versioned files so an environment can be created, reviewed and rebuilt automatically.
A compact, signed token that carries claims about a user, so a server can verify who is calling without looking up a session.
A system that runs and scales containers across many servers and restarts them if they fail.
A framework for building applications around language models, with building blocks for prompts, tools, memory and retrieval.
An open-source platform for real-time audio and video, often used to build voice agents.
A neural network trained on large amounts of text that can read and write language, answer questions and write code. GPT, Claude and Gemini are examples.
Coordinating prompts, tools, models and data sources across the steps of an AI workflow, including retries and fallbacks.
Tools that build applications or automations mostly through visual interfaces instead of code. Fast for simple needs, with limits as logic or scale grows.
A visual automation platform, formerly Integromat, that connects apps through scenarios.
An open standard that lets AI applications connect to tools and data sources in a consistent way, so one integration works across different models and clients.
An architecture that splits an application into small services that each do one job and talk over APIs. Easier to scale in parts, harder to operate than one application.
Microsoft's cloud platform for computing, storage, databases and AI services.
At Techdome: Sparrow is listed on the Microsoft Azure Marketplace.
A long-established protocol for talking to industrial devices such as sensors and controllers.
One repository holding several projects or packages, so code can be shared and changed together.
A lightweight publish-subscribe messaging protocol widely used for IoT devices and sensors.
One application serving many customers while keeping each customer's data separate, usually by schema or row-level rules.
At Techdome: Our telehealth platform runs 38 clinic tenants on a PostgreSQL schema with full tenant data isolation and role-based access.
The smallest version of a product that real users can use, built to find out whether the idea works before investing more.
At Techdome: Our Blueprint takes a product from a day-zero repository to first paying customers in eight weeks, and fixed-scope sprints of two to four weeks start at $8,000.
A workflow automation platform that connects apps and APIs through visual workflows and can be self-hosted on your own servers.
At Techdome: For a US mortgage lender, n8n workflows automated 80% of back-office document processing. Our automation sprints take two to four days to a working production flow.
A way of running n8n where a main instance hands executions to separate worker processes through Redis, so many workflows run in parallel and the system scales.
At Techdome: We run self-hosted n8n clusters in queue mode with Redis for production clients.
Building software with a team in a nearby country or time zone, which makes real-time collaboration easier, usually at a higher cost than offshore.
A graph database that stores data as nodes and relationships and is queried with the Cypher language.
A TypeScript framework for Node.js servers that adds structure, modules and dependency injection, popular for larger back-end applications.
A React framework that adds routing, server rendering and API routes, for fast, search-friendly websites and apps.
At Techdome: This website is built with Next.js.
A runtime that runs JavaScript on the server, commonly used for APIs and real-time services.
At Techdome: The backends of our telehealth platform and automation sprints run on Node.js.
An inference server that serves machine learning models from many frameworks on GPUs and CPUs.
OAuth 2.0 lets an app act on a user's behalf with limited access; OpenID Connect adds a standard way to prove who the user is. Together they power most 'sign in with' flows.
Understanding what a system is doing from its logs, metrics and traces, so problems can be found and explained quickly.
Software that turns images of text, such as scanned documents, into machine-readable text.
At Techdome: Our mortgage lender automation reads documents with OCR and cut manual back-office work by 80%.
Building software with a team in a distant country, often at lower cost. The main thing to plan for is time-zone overlap.
At Techdome: Our hubs span the US, UAE and India, with four to six hours of daily overlap with US and European teams.
A tool for running open-weight language models on your own machine or server.
An industrial communication standard for exchanging data between machines, sensors and software.
A family of large language models from OpenAI, available through an API and used for chat, writing, coding and structured extraction.
An open-source library for computer vision: reading images and video, detecting objects and processing frames.
Replacing a card number with a random token that is useless if stolen, so the real number stays with the payment provider.
The security standard for any system that stores, processes or transmits payment card data.
At Techdome: We design payment flows so raw card numbers never reach our backend servers, using tokenised processing.
The ISO standard for accessible PDF documents, so screen readers can read their structure.
A managed vector database service for similarity search over embeddings.
An open-source relational database known for reliability, strong SQL support and extensions such as pgvector for AI search.
At Techdome: PostgreSQL runs our multi-tenant telehealth platform and our GraphRAG systems.
A TypeScript-friendly ORM that maps database tables to typed objects and manages schema migrations.
When a product meets a real market need strongly enough that customers keep using it and recommend it.
A web app that can be installed, work offline and send notifications, giving an app-like experience from a browser.
At Techdome: We re-platformed a clinic operating system as a PWA mid-flight.
Writing and testing the instructions, examples and context given to a language model so it returns reliable output for a task.
A small build that tests whether something can be done technically, before committing to a full product. A POC asks 'can we?', where an MVP asks 'do users want it?'.
A general-purpose programming language widely used for web back ends, data work and machine learning.
An open-source machine learning framework, widely used in research and for training deep learning models.
Gives a language model the most relevant documents at question time, so its answers are grounded in your own data and can cite it.
A cloud platform for deploying apps and databases from a repository with little infrastructure setup.
Giving users permissions according to their role, so people can only see and change what their job requires.
A JavaScript library for building user interfaces from reusable components, widely used for web applications.
At Techdome: Our senior frontend engineers build in React and Next.js.
An in-memory data store used for caching, queues and fast lookups.
Re-running tests after a change to confirm that features which worked before still work.
A common way to build APIs over HTTP, using URLs for resources and standard methods such as GET and POST.
A platform for building voice AI agents that make and take phone calls.
Software bots that imitate clicks and keystrokes in an application's screens. Useful when a system has no API, and more fragile than API-based automation.
A systems programming language focused on speed and memory safety.
At Techdome: Sparrow is built with Rust and TypeScript.
Software sold by subscription and run by the vendor in the cloud, so customers use it in a browser without installing or hosting it.
A cloud CRM platform for sales, service and customer data, with a large ecosystem of integrations.
An XML-based standard for single sign-on between a company's identity provider and the applications its staff use.
Code added to a page that describes its content in a standard vocabulary, so search engines can show rich results and understand who and what the page is about.
Software that reads screen content aloud or to braille for blind and low-vision users. NVDA and JAWS run on Windows, VoiceOver on Apple devices.
SFTP moves files over an encrypted connection. EDI exchanges standard business documents, such as orders and invoices, between companies' systems.
The standard signalling protocol for voice calls over IP, used to connect phone systems and carriers.
An audit report on a company's controls for security, availability and confidentiality, often requested by business customers of software vendors.
A document that sets out what will be delivered, the timeline, the cost and how acceptance works for a project.
Adding external engineers to your team for a project or period, with you managing their work day to day.
At Techdome: We embed a pod in about two weeks, with four to six hours of daily overlap with US and European time zones and a 14-day replacement guarantee.
Turning text into audio in small chunks as it is generated, so a voice agent can start speaking before the whole reply is written.
A payment platform with APIs for cards, subscriptions and invoicing, which keeps card data on its own servers instead of yours.
An open-source backend platform built on PostgreSQL, with authentication, storage and real-time APIs.
A utility-first CSS framework: you style elements with small single-purpose classes in the markup instead of writing separate stylesheets.
A founder who leads engineering and shares ownership of the company. Founders without one often use a fractional CTO or a studio to build the first version.
The future cost of shortcuts taken in code or design to ship sooner. Like financial debt, it is fine if repaid on purpose and harmful if ignored.
At Techdome: Our studio architecture work designs schemas, tenant isolation and failover before feature code, to keep debt out from day one.
Assessing an engineer's skills with a structured test or interview before hiring.
At Techdome: We built JustInterview.ai to screen engineers, and used it to hire our own 110-person team.
Google's open-source machine learning framework for training and deploying models.
A tool for defining cloud infrastructure in code and applying it repeatably.
A contract billed on the hours and resources used. Flexible when scope will change, with less price certainty than fixed price.
A communications platform with APIs for voice calls, SMS and messaging.
JavaScript with types. It catches many mistakes before code runs and makes large codebases easier to change.
Stores embeddings and finds the closest matches by meaning rather than keyword. Many teams use a vector extension in the database they already run.
At Techdome: We use PostgreSQL with the pgvector extension in our GraphRAG systems.
A company that builds and launches software products and ventures in-house, sharing its engineers and playbook across each one.
At Techdome: Techdome builds its own products (Sparrow, JustInterview.ai, Aether Voice and Techdome ARC) with the same pods and Blueprint it uses for client ventures.
A high-throughput inference engine for serving large language models efficiently on GPUs.
An AI agent that talks over a phone call or voice interface in real time. It needs fast speech recognition and synthesis to feel natural.
At Techdome: Aether Voice is our voice agent product: it answers and qualifies calls with memory and calendar sync.
The Web Content Accessibility Guidelines: the international standard for making websites and apps usable by people with disabilities. Level AA is the usual legal and procurement target.
At Techdome: For a UAE public-sector tourism programme we audited against WCAG 2.2 AA and EN 301 549, with about 640 hours of testing.
An HTTP request one system sends to another when an event happens, so the second system reacts at once instead of repeatedly checking.
At Techdome: Our own publishing flow uses a webhook: publishing a post in our CMS triggers the subscriber email, with a check that each post is sent only once.
A browser standard for real-time audio, video and data between peers, with low latency.
A protocol that keeps a two-way connection open between a browser and a server, so data can be pushed in real time without repeated requests.
Meta's hosted API for sending and receiving WhatsApp messages from business systems.
Using software to run repeatable business processes, such as approvals, data entry and notifications, without manual steps.
At Techdome: Every automation sprint we ship has automatic retries and alerting, and our five recent sprints ran at 99.9% reliability.
'You only look once': a family of fast object-detection models that find and label objects in an image in a single pass.
A hosted automation service that connects apps through triggers and actions, priced by the number of tasks it runs.