Amazon Q Developer Is Being Retired: How to Track Usage Through the Kiro Migration
AWS gave Amazon Q Developer customers until April 30, 2027 before the IDE plugins for VS Code, JetBrains, Eclipse, and Visual Studio stop working entirely. New signups already closed on May 15, 2026. If your team has Q Developer seats today, you're on a migration clock whether anyone has said so out loud yet.
This isn't a small vendor sunsetting a side project. Q Developer was AWS's answer to GitHub Copilot, and its replacement, Kiro, changes the interaction model from inline suggestions to spec-driven agentic workflows. That means a re-onboarding, not a version bump, for every developer who has to switch.
TL;DR
- Amazon Q Developer's IDE plugins and paid subscriptions end on April 30, 2027; new signups have been blocked since May 15, 2026, and Opus 4.6 stopped shipping to Q Developer Pro on May 29, 2026.
- AWS's recommended path is Kiro, a spec-driven agentic IDE with a dashboard structure almost identical to Q Developer's: subscription counts, active users, and credits consumed, updated hourly.
- Kiro's dashboard has the same blind spots Q Developer had: no cross-tool comparison, no commit or PR correlation, and no automated ghost-seat alerting beyond a raw "pending subscriptions" count.
- Usage history built up inside Q Developer's console does not follow you into Kiro or any other tool. Export it, or run tracking that sits outside any single vendor's dashboard, before the 2027 deadline.
- A forced migration is also a natural point to re-audit seats. Teams that skip this step tend to re-provision the same headcount into the new tool without checking who was active in the old one.
What is happening to Amazon Q Developer, and by when?
AWS is retiring Amazon Q Developer in stages: new signups blocked since May 15, 2026, model updates frozen starting May 29, 2026, and full end of support on April 30, 2027.
| Date | What changes |
|---|---|
| May 15, 2026 | New free-tier accounts and new paid subscriptions blocked. Existing Pro subscriptions can still add users. |
| May 29, 2026 | Opus 4.6 removed from Q Developer Pro. Opus 4.5 and earlier models stay available. |
| April 30, 2027 | Full end of support. IDE plugins for VS Code, JetBrains, Eclipse, and Visual Studio stop functioning. Code transformation features (Java upgrades, .NET porting) go with them. |
Q Developer inside the AWS Management Console, AWS documentation search, the Console Mobile App, and the Slack/Teams integrations are not affected. This retirement is specifically about the IDE-embedded coding assistant, the part your developers use every day.
The clock is longer than it looks. April 30, 2027 sounds distant, but AWS Transform migrations (the Java and .NET modernization tooling tied to Q Developer) have their own recommended completion windows before that date. Treat 2027 as the hard deadline, not the planning deadline.
What does Kiro's usage dashboard show that Amazon Q Developer's didn't?
Almost nothing new. Kiro's enterprise dashboard tracks subscriptions, active users, and credit consumption in nearly the same shape as Q Developer's, because AWS built it on the same underlying telemetry pipeline.
| Capability | Amazon Q Developer dashboard | Kiro enterprise dashboard |
|---|---|---|
| Total / active / pending subscriptions | Yes, org-wide with member-account rollup | Yes, broken out per tier (Pro, Pro+, Pro Max, Power) |
| Active users over time | Daily line graph | Daily/weekly/monthly, split by IDE vs. CLI |
| Feature-level acceptance rates | Yes (inline suggestions, chat, doc gen, unit tests) | Not part of the public dashboard docs |
| Credits or usage-based spend | Lines of code by feature (pie chart) | Credits consumed, by tier and client type |
| Per-user daily CSV export | Yes, via S3, one file per 1,000 users | Yes, via S3, split by client type |
| Cross-tool comparison (vs. Copilot, Claude Code, Cursor) | No | No |
| Ghost-seat alerting | Only a raw "pending subscriptions" count, no trend or alert | Only a raw "pending subscriptions" count, no trend or alert |
| Commit or PR-level correlation | No | No |
Both dashboards update on an hourly cycle for most metrics, with the active-users widget refreshing daily on UTC. Both offer a per-user activity export as a scheduled CSV drop into an S3 bucket rather than a queryable API. If your team built any process around pulling Q Developer's CSVs, that same process needs to point at a new bucket path for Kiro, not a new data model.
What should engineering teams do before the deadlines hit?
Treat this as a standard vendor-migration project: audit current usage, decide on a destination tool, export historical data, and pilot before reassigning every seat.
- Pull your Q Developer usage export now. The per-user activity CSVs live in S3 for as long as you keep collecting them, not indefinitely by default. Archive at least 90 days of history before you decommission the reporting pipeline.
- Decide if Kiro is the right destination. Staying inside AWS IAM Identity Center is convenient if your org is already AWS-committed. It is not the only option: GitHub Copilot, Cursor, and Claude Code accomplish the same job with different admin models.
- Re-audit seats before re-provisioning. We found 24% of seats in a typical AI coding tool rollout were never activated — a migration is the cheapest moment to drop dead seats, because nobody has to notice you removed something they weren't using anyway.
- Run a 30-day pilot on the new tool with the same metrics you tracked before. Adoption rate, active-seat ratio, and time-to-first-session transfer cleanly across any AI coding tool. Don't let the migration also become the moment you stop measuring.
- Keep monitoring during the cutover. Broken installs and silent setup failures are common during any tool switch, and they look identical to low adoption unless something is watching for them specifically.
How do you avoid losing a year of usage data in the switch?
Export Q Developer's historical CSVs before decommissioning, and put any new adoption baseline in a system that isn't tied to the tool you might switch away from again.
This is the part most migration guides skip. AWS's own dashboards, for both Q Developer and Kiro, live entirely inside that vendor's console. If your only record of "seat X went quiet in March" sits inside Q Developer's per-user report, that record does not follow the seat when it becomes a Kiro seat, or a Copilot seat, or anything else.
Teams that ran a vendor-agnostic tracking layer alongside Q Developer keep their adoption history, ghost-seat trend, and per-developer baseline intact through the switch. Teams that relied solely on the native dashboard start over at zero on day one of the new tool, with no way to tell whether a quiet developer went quiet before or after the migration.
This is the same lock-in problem, one layer up. BYOK exists so switching model providers doesn't mean losing your billing history; the same principle applies to switching the IDE tool itself. The console you're renting is not the place to keep the data you need to survive changing consoles.
FAQ: Amazon Q Developer retirement and migration
When does Amazon Q Developer stop working?
Amazon Q Developer IDE plugins and paid subscriptions reach full end of support on April 30, 2027. New signups (free and Pro) were already blocked on May 15, 2026, though existing Pro subscriptions can still add users until the 2027 cutoff.
What is replacing Amazon Q Developer?
AWS is directing customers to Kiro, a spec-driven agentic development environment built for planning, implementing, and verifying changes from structured specifications rather than reactive inline completions. Kiro carries the newest Claude models; Q Developer Pro stopped receiving Opus 4.6 on May 29, 2026.
Does Kiro's admin dashboard show the same data as Amazon Q Developer's?
Mostly yes. Kiro's enterprise dashboard tracks total, active, and pending subscriptions per tier, active users by client type, and credits consumed, updated hourly except for the daily active-users widget, which mirrors Q Developer's structure closely. Neither tool shows cross-tool comparisons, commit-level impact, or delivery-metric correlation.
What happens to our Amazon Q Developer usage history when we migrate?
It stays in AWS's per-account dashboard and CSV exports, but it does not carry over into Kiro or any other tool automatically. If your adoption baseline, ghost-seat trend, or productivity data lived only inside Q Developer's console, migration resets that history to zero unless you exported it or ran a tool-independent tracking layer.
Should we migrate to Kiro or switch to a different AI coding tool entirely?
Treat it as a genuine build vs. buy decision, not a forced upgrade. Kiro keeps you in the AWS IAM Identity Center ecosystem with a similar admin model; GitHub Copilot, Cursor, and Claude Code are equally valid alternatives if your team isn't AWS-committed. Run a 30-day pilot with the same adoption and ghost-seat metrics you already track before committing seats.
Related reading
- 24% of our AI licences were never activated, the seat-audit playbook to run before you re-provision anyone into a new tool
- How to track GitHub Copilot usage across your engineering team, a native-analytics breakdown for the most common Q Developer replacement
- How to track Cursor usage and ROI across your engineering team, the same adoption-tracking approach for another common destination
- BYOK for Anthropic Claude, why vendor lock-in at the billing layer causes the same problem this migration does