New blog post: ai-based code review by TomasTomecek · Pull Request #1107 · packit/packit.dev

TomasTomecek · 2026-02-19T16:45:03Z

You are very welcome to share your perspective with AI-powered code review because I tried to write it as it's a team effort.

centosinfra-prod-github-app · 2026-02-19T16:48:25Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/9f29b5fb91cf416da300bcc1bde12ece

✔️ pre-commit SUCCESS in 1m 04s

lbarcziova

nice work!

posts/code-review/index.md

lbarcziova · 2026-02-24T09:01:19Z

posts/code-review/index.md

+
+## [Gemini-code-assist](https://github.com/marketplace/gemini-code-assist)
+
+Our AI tool of choice for our upstream reviews is Gemini-code-assist. It's free, easy to set up, and you get to see the comments inside the GitHub interface. Since it's powered by the Gemini models, it can provide solid feedback. If your PR contains clear mistakes, this tool can easily find them.


If your PR contains clear mistakes, this tool can easily find them.

My feeling is that we revealed with it not only clear mistakes, but also more subtle ones, and it definitely prevented lot of bugs from getting into production. But curious what do others think.

Agreed, sometimes it can find very non-obvious issues.

amazing feedback, I'll incorporate it

lbarcziova · 2026-02-24T09:05:00Z

posts/code-review/index.md

+
+![Gemini referencing the style guide](img/gemini_ref.png)
+
+I'd say the biggest downside of this tool is that you need to open a PR to see the feedback, you cannot run it locally.


I would maybe also mention the related advantage, that it can encourage the team to align on technical debt or complex logic that the tool brings up as a review comment.

centosinfra-prod-github-app · 2026-02-24T09:27:40Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/24e7235747f14896920fb48a67503706

✔️ pre-commit SUCCESS in 1m 07s

posts/code-review/index.md

nforro · 2026-02-24T09:31:35Z

posts/code-review/index.md

+
+This tool can be infinitely helpful, especially if your codebase is easy to navigate, is not colossal, and has solid documentation which Claude can follow.
+
+Several people on the team use Claude Code daily and it is part of our daily job. We've gotten used to it so much that we couldn't live without it, honestly. It's so convenient and helpful.


I'm yet to give it a try 🙈

oh true, let me add that info there

centosinfra-prod-github-app · 2026-02-24T09:33:36Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/186ffc6c4aeb41bdbf09b92048d917ac

✔️ pre-commit SUCCESS in 1m 08s

centosinfra-prod-github-app · 2026-02-24T09:44:27Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/c283632425fe4aef9e1c5e106cc47be8

✔️ pre-commit SUCCESS in 1m 06s

For sake of running script `files/generate_usage.py` Signed-off-by: Tomas Tomecek <ttomecek@redhat.com>

Signed-off-by: Tomas Tomecek <ttomecek@redhat.com> Assisted-by: Claude

centosinfra-prod-github-app · 2026-02-24T09:48:32Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/1df80a2c79d74181babcd9153b7b4964

✔️ pre-commit SUCCESS in 1m 05s

lbarcziova

🚀

centosinfra-prod-github-app · 2026-02-24T10:19:57Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/9801910cc10443b6adc5abb23a8ca7af

✔️ pre-commit SUCCESS in 1m 05s

posts/code-review/index.md

mfocko · 2026-02-24T10:31:27Z

posts/code-review/index.md

+
+# Our experience with AI-based code review
+
+One amazing benefit that modern LLMs come with is using them as a linter, or a pair programmer. You can easily get feedback on your code with such an approach. If the feedback is solid, your code is improved. If the review is poor, you can just disregard it. But overall with very little effort you can gain a lot.


You can easily get feedback on your code with such an approach.

Is this supposed to be interpreted as using the LLM?

you can just disregard it

or prompt again…

improved the first paragraph to address the first comment, but second is silently implied, we don't need to be pixel-perfect here :)

posts/code-review/index.md

centosinfra-prod-github-app · 2026-02-24T13:00:40Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/b7845daedaf34682910c558dc92db640

✔️ pre-commit SUCCESS in 1m 10s

Co-authored-by: Laura Barcziová <49026743+lbarcziova@users.noreply.github.com> Co-authored-by: Nikola Forró <nforro@redhat.com> Co-authored-by: Maja Massarini <mmassari@redhat.com> Co-authored-by: Matej Focko <mfocko@users.noreply.github.com> Signed-off-by: Tomas Tomecek <ttomecek@redhat.com> Assisted-by: Claude

centosinfra-prod-github-app · 2026-02-24T13:13:17Z

Build succeeded.
https://gateway-cloud-softwarefactory.apps.ocp.cloud.ci.centos.org/zuul/t/packit-service/buildset/9f8d31b866d1484ba991214587bb72ef

✔️ pre-commit SUCCESS in 1m 06s

usercont-release-bot added this to Packit pull requests Feb 19, 2026

github-project-automation bot moved this to New in Packit pull requests Feb 19, 2026

TomasTomecek force-pushed the code-review branch from 30a939b to f9db957 Compare February 19, 2026 16:46

lbarcziova reviewed Feb 24, 2026

View reviewed changes

nforro reviewed Feb 24, 2026

View reviewed changes

posts/code-review/index.md Outdated Show resolved Hide resolved

nforro reviewed Feb 24, 2026

View reviewed changes

TomasTomecek force-pushed the code-review branch from 36c86a3 to 12181af Compare February 24, 2026 09:42

TomasTomecek added 2 commits February 24, 2026 10:46

add requirements.txt

ada0118

For sake of running script `files/generate_usage.py` Signed-off-by: Tomas Tomecek <ttomecek@redhat.com>

add blog for AI-powered code review

10d0ed7

Signed-off-by: Tomas Tomecek <ttomecek@redhat.com> Assisted-by: Claude

TomasTomecek force-pushed the code-review branch from 12181af to 198efd7 Compare February 24, 2026 09:46

lbarcziova approved these changes Feb 24, 2026

View reviewed changes

TomasTomecek force-pushed the code-review branch from 198efd7 to 5814f92 Compare February 24, 2026 10:18

TomasTomecek added the mergeit Merge via Zuul label Feb 24, 2026

mfocko reviewed Feb 24, 2026

View reviewed changes

TomasTomecek removed the mergeit Merge via Zuul label Feb 24, 2026

TomasTomecek force-pushed the code-review branch from adf81b3 to c43f53f Compare February 24, 2026 13:11

mfocko approved these changes Feb 24, 2026

View reviewed changes

TomasTomecek merged commit 37961e7 into packit:main Feb 24, 2026
4 checks passed

github-project-automation bot moved this from New to Done in Packit pull requests Feb 24, 2026

TomasTomecek deleted the code-review branch February 24, 2026 14:35


		## [Gemini-code-assist](https://github.com/marketplace/gemini-code-assist)

		Our AI tool of choice for our upstream reviews is Gemini-code-assist. It's free, easy to set up, and you get to see the comments inside the GitHub interface. Since it's powered by the Gemini models, it can provide solid feedback. If your PR contains clear mistakes, this tool can easily find them.


		![Gemini referencing the style guide](img/gemini_ref.png)

		I'd say the biggest downside of this tool is that you need to open a PR to see the feedback, you cannot run it locally.


		This tool can be infinitely helpful, especially if your codebase is easy to navigate, is not colossal, and has solid documentation which Claude can follow.

		Several people on the team use Claude Code daily and it is part of our daily job. We've gotten used to it so much that we couldn't live without it, honestly. It's so convenient and helpful.


		# Our experience with AI-based code review

		One amazing benefit that modern LLMs come with is using them as a linter, or a pair programmer. You can easily get feedback on your code with such an approach. If the feedback is solid, your code is improved. If the review is poor, you can just disregard it. But overall with very little effort you can gain a lot.

Conversation

TomasTomecek commented Feb 19, 2026

Uh oh!

centosinfra-prod-github-app bot commented Feb 19, 2026

Uh oh!

lbarcziova left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

lbarcziova Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

nforro Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

TomasTomecek Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

lbarcziova Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

Uh oh!

nforro Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

TomasTomecek Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

lbarcziova left a comment

Choose a reason for hiding this comment

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

Uh oh!

mfocko Feb 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

TomasTomecek Feb 24, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

centosinfra-prod-github-app bot commented Feb 24, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

mfocko Feb 24, 2026 •

edited

Loading