Release notes automation and a model change log you can stand behind
Models, system prompts, safety filters and policies change often, and customers notice behavior changes before they read about them. We build a workflow that collects merged pull requests, model and prompt version changes and evaluation results, drafts customer-facing release notes and an internal change log, routes them for approval and publishes to your changelog, docs and Slack. A typical build costs $1,000 to $2,500 and takes about four weeks.
The problem
Why AI product changes are hard to explain
An AI product changes in more places than a normal app: code, model version, system prompt, retrieval settings, safety filters, usage policy. Each change can shift behavior, and customers ask why outputs look different this week.
Without a reliable record, explaining a change to a customer, answering an auditor or simply proving what was deployed on a given date means digging through commits, chat threads and memory.
What we build
What our release notes automation includes
Change collection
Merged pull requests, deployment events, model and prompt version changes and policy edits collected from GitHub, GitLab or Linear.
Eval results attached
Each model or prompt change linked to its evaluation run, so the change log shows the evidence.
Drafted release notes
Customer-facing notes written in your style from the changes, grouped by product area.
Internal change log
A detailed record of what changed, why, who approved it and when it shipped.
Approval step
Product and model owners approve notes before they are published.
Publishing
Approved notes posted to your changelog page, docs and Slack or email.
Who does what
Automation does the chasing, people make the calls.
-
01
Automated
Changes collected
Merged changes, deploys and version changes are collected for the release window.
-
02
Automated
Evidence linked
Evaluation runs and approvals are linked to each model or prompt change.
-
03
Automated
Notes drafted
Customer release notes and the internal change log are drafted.
-
04
Team review
Owners review
Product and engineering owners edit and approve the drafts.
-
05
Founder or model owner decides
Disclosure decisions
What to disclose about behavior and policy changes is decided by the model owner or founder.
How it behaves
How the model change log behaves
What customers see
- Clear release notes when behavior changes
- A changelog they can subscribe to
- Consistent explanations from support
What the team sees
- Drafted notes and a change log entry for each release
- Eval evidence linked to model and prompt changes
- A simple approval step
What the automation does
- Collects changes and version events for the window
- Drafts notes and change log entries
- Publishes only after approval
Edge cases and fallbacks
- Change with no description
- Listed for the owner to describe; no note is guessed.
- Security fix that should not be detailed
- Labelled and summarised generically for the owner to decide.
- Model change with no eval run
- Flagged as missing evidence before approval.
- GitHub or n8n unavailable
- The draft is delayed and the owner is told; the team can publish manually.
Architecture
Release notes and change log at a glance.
- Input
- Logic and services
- Data
- AI step
- People
- Output
How it flows
- 1 Merged changes, deploys and model, prompt or policy version changes are collected for the release window.
- 2 Evaluation results are linked to each model or prompt change, and drafts are written in your style.
- 3 Product and model owners edit and approve the notes and the internal change log.
- 4 Approved notes are published to the changelog, docs and Slack.
Project planner
How the project runs, phase by phase.
Release notes automation & model change log project plan
Typical time to launch: about 4 weeks, then support
-
01
Discovery
2 to 3 days
Release rhythm, where changes are recorded, style guide and who approves.
-
02
Design
2 to 3 days
Change sources, labels, note templates, change log fields and publishing targets.
-
03
Build
1 to 2 weeks
Collection, evaluation linking, drafting, approval and publishing.
-
04
Shadow run
About 1 week
Drafts produced for a release while the team writes notes as usual, and we compare.
-
05
Launch
1 to 2 days
Use for the next release with owners approving.
-
06
Support
Ongoing
Template changes, new sources and publishing targets.
What it runs on
Real products only, connected through the access each vendor provides. See all integrations
Cost estimate
What it costs, line by line.
-
Discovery and design
Sources, templates and approvals.
$200 – $400
-
Build
Collection, evidence linking, drafting and publishing.
$500 – $1,300
-
Integrations
Git host, issue tracker, eval store and docs site.
$150 – $400
-
Testing and shadow run
Drafts compared with hand-written notes.
$100 – $250
-
Launch and training
First release and owner walkthrough.
$50 – $150
-
Total project estimate
$1,000 – $2,500
Optional monthly care after launch is quoted with your proposal. Fixed quote after a free call. Model and n8n costs are billed to you directly.
Monthly running costs you pay directly
Estimates from the vendors' published pricing, checked November 2026. Billed to you by each vendor, never marked up.
- n8n: Free self-hosted Community edition, or n8n Cloud from about €20 a month (Starter) or €50 (Pro), billed annually.
- Model usage: Billed by the provider per use; small per release.
What pushes cost up
- Systems without an API, or with integration access that must be requested from the vendor
- Writing data back into your core system rather than only reading from it
- Several locations, brands or very different processes to support
- Custom apps, dashboards or a voice line on top of the workflows
What keeps cost down
- One location and one clear process to start with
- Cloud software with a documented API
- Using tools you already pay for instead of adding new ones
- Approved wording and sample data ready at the start
Privacy and consent
Rules we design for, with the sources.
EU: documentation for model providers
Article 53 requires general-purpose model providers to keep technical documentation up to date and give downstream providers information about capabilities and limitations; a change log supports both.
Source: artificialintelligenceact.euUS: govern and manage
The NIST AI RMF, used voluntarily, emphasises documented governance and management of AI changes over time.
Source: www.nist.govEU: AI disclosure obligations
Where your product falls under Article 50 transparency duties from 2 August 2026, the change log records when disclosures changed.
Source: artificialintelligenceact.euFree tools
Try these first, free.
How we deliver
Agile delivery, in 1 to 2 week sprints.
Every sprint
- 1 Sprint planning. At the start of each sprint we agree which backlog items we will finish, each with written acceptance criteria you can check.
- 2 Daily stand-up. A short internal check-in on progress and blockers. You get a written update in your Slack or Teams channel, not another meeting.
- 3 Sprint review and demo. At the end of each sprint we demo what was built on test data or in a shadow run, and you accept it or ask for changes.
- 4 Retrospective. The team looks at what went well and what to change in the next sprint, and we share anything that affects you.
How scope stays under control
- Product backlog
- Every feature, fix and question in one prioritised list you can see at any time.
- Acceptance criteria
- Plain-language conditions for "done" on each item, agreed before work starts.
- Change requests
- New ideas go into the backlog with an estimate. Nothing changes scope or price without your written approval.
- User acceptance testing (UAT)
- Before launch your team tests the real workflow against the acceptance criteria and signs off.
Tools we use
Project manager / Scrum master
Runs the sprints, keeps the backlog and timeline honest, and is your single point of contact.
Business analyst
Maps how the work runs today, writes the workflow specification and acceptance criteria with your team.
Automation engineer
Builds the n8n workflows, integrations, AI steps and error handling.
Backend developer
Writes custom code where a no-code step is not enough: APIs, data sync, secure storage.
Frontend / mobile developer
Builds forms, booking pages, dashboards, web apps or Android apps when the project needs them.
QA engineer
Tests every workflow against the acceptance criteria, including edge cases and failure paths.
DevOps engineer
Sets up hosting, access, backups, monitoring and alerts, and the deployment pipeline.
Who does what at each phase
- Free call
- Project manager, business analyst
- Understand the problem, check it is worth automating and outline a rough range.
- Discovery
- Business analyst, automation engineer
- Map the current process, systems and data, and agree success measures.
- Design
- Business analyst, automation engineer, backend developer
- Write the workflow specification, data flows, compliance notes and backlog with acceptance criteria.
- Sprints
- Automation engineer, developers, QA
- Build, test and demo in 1 to 2 week sprints; you accept or adjust after each demo.
- UAT and launch
- QA, DevOps, project manager
- Your team tests and signs off, we switch on, monitor closely and train the team.
- Support
- Automation engineer, DevOps
- Stabilisation window, then optional monthly care with monitoring and changes.
A point of contact
One person who can answer questions and approve decisions, with about an hour a week during the project.
Access
Admin or API access to the systems involved, created as separate accounts with the least access needed.
Sample data
Real but non-sensitive records (or de-identified ones) so we can design and test against reality.
Approved wording
Message templates, policies and consent wording (for SMS, your opt-in and opt-out text) reviewed by you.
Vendor agreements
Signed agreements with the software vendors that will handle personal data, such as a BAA or processor contract.
Time for testing
A few hours from the people who will use the system, for the demos and user acceptance testing.
Working system
Workflows, integrations and any apps running in your accounts, under your keys.
Documentation
Workflow specification, data-flow map, run book and exported workflow files.
Training
A walkthrough for the team and a short recorded video for new starters.
Handover
All access, credentials and code handed over; you can keep us or take it in-house.
Stabilisation window
Fixes for anything that does not meet the agreed acceptance criteria after launch.
Support options after launch
No plan
You run it yourselves after the stabilisation window and contact us when you need a change.
Monthly care
Monitoring, error alerts, small changes and updates each month, with an agreed response time.
Priority care
Faster response times for business-critical workflows and a monthly review call.
How a project moves through our team
- Input
- Logic and services
- People
- Output
- Data
How it flows
- 1 A free call with the project manager and business analyst to understand the problem and whether it is worth building.
- 2 Discovery and design produce a written specification, data-flow map and a backlog with acceptance criteria.
- 3 The team builds in 1 to 2 week sprints; you review a demo at the end of each and accept or adjust.
- 4 Your team runs user acceptance testing, we launch, then monitor and improve under a support plan.
FAQ
Release notes automation & model change log questions, answered.
The workflow reads merged pull requests and their labels through the GitHub API for the release window, groups them by product area and drafts notes in your style. Owners approve before anything is published.
Yes. Linear has a GraphQL API with webhooks, and Jira has a REST API. Issues linked to merged changes add context to the notes.
An internal record of every change to model versions, system prompts, safety settings and policies, with the evaluation evidence and who approved it. It answers "what was live on this date and why".
No. Drafts wait for owner approval, and decisions about what to disclose regarding behavior changes stay with the model owner or founder.
Yes. Approved notes can be sent by email or RSS from your changelog page.
Time from release to published notes, releases with complete change log entries and customer tickets asking why behavior changed.
Free 20-minute call
Want this for your ai company?
Book a free 20-minute call. We will look at how the work runs today, tell you honestly whether it is worth building and give you a fixed quote if it is.
More for ai companies
All servicesCRM for AI startups: signups to sales pipeline
Route self-serve signups, demo requests and API key creations into one CRM, score accounts on real product usage, alert sales to product-qualified leads and keep founders out of spreadsheet triage.
From $1,000 · About 4 weeks Details
LLM evaluation pipeline & regression testing
Run your evaluation set on every prompt, model or retrieval change in CI, compare scores with the last release, block regressions and post a readable summary to the pull request and Slack.
From $1,000 · About 4 to 5 weeks Details
AI support agent for API and developer docs
An AI support agent that answers developers from your current docs, changelog and status page, explains errors and rate limits, and opens a well-formed ticket with request IDs when a person is needed.
From $2,500 · About 5 to 8 weeks Details