AI glossary Operations
What is a technical specification (tech spec)?
Technical Specification (Tech Spec)
← All glossary termsAn engineering blueprint that describes how a system will be built: services, data models, APIs, infrastructure, risks and rollout plan.
Explained
A technical specification, or tech spec, turns what a system must do into how it will be built. It describes the architecture and components, the data models, the APIs and integrations, the infrastructure, the non-functional targets such as performance and availability, the security measures, and the migration and rollout plan, together with the trade-offs considered. It is written before the main build so that problems surface while they are still cheap to fix and the team agrees on one design. For AI systems it also covers the choice of model, evaluation suites, fallback paths and observability, so that tools and agents built on top have a stable contract to target.
What a tech spec usually contains
- Context and goals, with a link to the functional requirements.
- Proposed architecture: components, responsibilities and how they interact.
- Data model and storage.
- APIs, events and integrations with other systems.
- Non-functional requirements: performance, availability, scalability and cost, often as service level objectives (SLOs).
- Security, privacy and compliance.
- Alternatives considered and why they were rejected.
- Migration, rollout, testing and monitoring plan.
- Risks and open questions.
Technical vs functional specification
A functional specification describes what the system must do from the user's and the business's point of view: features, rules and acceptance criteria. A technical specification describes how it will be done. The first is reviewed mainly by product and business owners; the second by architects and engineers. Both should trace to each other, so that every design decision answers a requirement.
What changes for AI systems
- Model choice and why: capability, cost, latency, hosting and data residency.
- Evaluation suites with the quality thresholds the system must meet before release.
- Fallback paths when the model fails, is slow or is unsure, including handover to a person.
- Guardrails, permissions and limits on the actions agents may take.
- Observability: logging of prompts, tool calls, cost and answer quality.
- How prompts, models and indexes will be versioned and rolled back.
Frequently asked questions
What is a tech spec in software development?
It is a document that explains how a feature or system will be built: architecture, data, APIs, infrastructure, risks and rollout. It is written before the main build so the team agrees on the design and catches problems early.
What is the difference between a technical and a functional specification?
The functional specification says what the system must do for users and the business. The technical specification says how engineering will build it. They complement each other and should be linked.
Is a tech spec the same as a design doc?
They are very close, and many teams use the terms interchangeably. A design doc tends to focus on the proposed design and its trade-offs; a tech spec may also include detailed contracts, targets and rollout plans.
Who writes the technical specification?
Usually the engineer or architect who will lead the work, with input from the rest of the team, security and operations. Product owners review it to check it meets the requirements.
What should a tech spec include for an AI system?
Besides the usual architecture, data and APIs, it should state the model choice, the evaluation criteria and thresholds, the fallback paths, the guardrails and permissions, and how quality and cost will be monitored in production.
Where it shows up at Thinkia
Further reading
Related terms
Sources
- Malte Ubl · Design docs at Google
- ISO · ISO/IEC/IEEE 29148:2018 Requirements engineering
- ISO · ISO/IEC/IEEE 42010:2022 Architecture description
Updated: 2026-10-21