Non-developer TPM Matias Singer built RADAR, an AI tool that turns stand-up transcripts into Jira tickets and weekly reports, cutting four hours to minutes.
You can turn meeting transcripts into Jira tickets by pasting the transcript into an AI tool that suggests action items, lets you review them and pushes the ones you approve to Jira. Matias Singer, a technical program manager at Jordan and Sam's company, Ryz Labs, built exactly that tool, called RADAR, and showed it on Episode 39 of Built This Week. He says it cut a four hour task down to minutes, and he is not a developer.
Matias joined five months before the episode as a TPM for Ntrvsta. Two months before recording, he started building with the help of AI, using Claude and Codex. The idea came from asking the team and developers what they were doing to automate their daily work, what was repetitive and how many tools they were juggling. RADAR was his attempt to gather all of that in one place.
Sam described the old workflow: sit in meetings, distill what was said, then create and assign every Jira ticket by hand. RADAR reduces that to a few clicks.
Watch RADAR turn a stand-up transcript into Jira tickets in Episode 39 (starts at 4:30) · Episode 39 notes and transcript
RADAR is available to every employee. It has sections for boards, analytics, transcripts, weekly business reviews (WBRs) and tail tickets. The projects in it are internal ones from teams like social, marketing and IT, and each board mirrors what you would see in Jira: open, in progress and done.
The analytics came from a practical problem. Matias says that in early meetings people were bored looking at a doc full of words, so he made it more dynamic with charts, which can make it easier to catch what matters.
Every morning the team has a stand-up with the engineers, recorded with Granola. Matias pastes that transcript into RADAR. In the demo the transcript was in Spanish, and RADAR identified five action items from it.
Sam pointed out the key design choice: RADAR does not push anything to Jira automatically, because not everything said in a meeting should become a ticket. Matias opens each suggestion, edits the title or details, assigns it to a project member, and selects or deselects items before creating them on the board.
Matias explained why this mattered so much to him:
So it was so annoying because if you were presenting, you were not able to type down what were the updates or a lot of times you miss things.
— Matias Singer, Episode 39
As a non-developer, he says he sometimes misheard or didn't follow what the engineers meant, and delays happened when he didn't raise a ticket in time. With RADAR, four hours of work now takes minutes. Asked about accuracy, he says it is pretty accurate, but a person still needs to oversee it.
Yes, and that is RADAR's other main feature. You upload one or several transcripts and it drafts the weekly business review. The report gives an overview, then breaks down each project with its subject, the important matters and the upcoming priorities. You can edit it, which was a feature Sam suggested the day before. Matias admits it is not always perfect and sometimes surfaces things that are not that important.
Features he was working on included adding tail tickets to the WBR, a pre-meeting section so teams can see where they left off the day before, storage for past transcripts, and a Granola integration so nobody has to paste a transcript at all. His goal: join the meeting, run it, and have the transcript, tickets and WBR handled afterward.
Sam says that over the previous three or four months the company pushed its whole team, from sales and marketing to TPMs and recruiters, to get comfortable building with AI. Step one was giving people the tools. The harder part was the cultural shift.
Matias described the format: a weekly AI meeting led by a senior developer, split into separate sessions for technical and non-technical people, where people share what they are building and how their work is changing.
People that never work with a developer, the sales team, marketing, finance team, they are all working on their own apps.
— Matias Singer, Episode 39
Sam added that it started with demo sessions walking non-technical people through very basic builds, then tool access, then a weekly show and tell. Almost every week someone presents a new tool. The accounting team has built financial dashboards for cash flow and AP and AR, and the social and sales teams have built their own tools too, often connected to Jira, Granola and Slack.
Jordan's answer is about ideas:
I think, like, it's impossible to have as many ideas as the crowd and the crowd is everyone who works for us.
— Jordan Metzner, Episode 39
People also learn from each other's builds and borrow features. Jordan is candid that this is still technical work, with back ends, front ends and GitHub involved, but he sees it getting closer to simply speaking to build the apps you need. The direction he wants is turning non-technical people technical, building tools that automate their own jobs, then scaling those tools to help everyone.
At the end of the episode Sam said he is proud that non-technical teammates have come to see AI with excitement rather than fear. Jordan called AI democratizing and a non-zero-sum game, and said Matias was proud enough of RADAR to show it to his girlfriend.
In the news, Sam covered Allbirds, the wool shoe brand that was everywhere in the 2010s. He says it sold last month for $39 million, far below a peak valuation he believed was over $1 billion, and then announced a pivot to AI. Jordan says the new company, to be called Newbird AI, plans to be a GPU as a service and AI native cloud provider. To him a shoe company making that move feels like peak cycle, and he doubts it will win much investor confidence. Sam compared it to the Long Island Iced Tea company pivoting to blockchain in 2017. We wished them luck.
Matias walks through RADAR's boards, the transcript to ticket flow and the WBR drafts on screen. Watch Episode 39 of Built This Week, where the full transcript is also on the page. For another internal tool built to save planning time, see Sam's offsite estimator in Episode 38.
