The founders behind Periscope, the livestreaming startup Twitter acquired in 2015, are back with a new venture — and this time, it’s all about AI.

 

On Wednesday, former Twitter head of product Kayvon Beykpour unveiled Macroscope, an AI system for developers and product leaders that automatically summarizes codebase changes, flags bugs, and streamlines collaboration.

 

Beykpour, now CEO, co-founded the company in July 2023 with longtime collaborator Joe Bernstein (also of Periscope and Terriblyclever, acquired by Blackboard in 2009) and Rob Bishop, who previously sold his machine learning startup Magic Pony Technology to Twitter in 2016.

 

Billed as an “AI-powered understanding engine,” Macroscope was born out of the founders’ frustrations at both startups and large organizations like Twitter. “Trying to get a sense of what everyone was doing was literally most of my job — and my least favorite part of it,” Beykpour told TechCrunch.

 

The platform integrates directly with GitHub (plus optional Slack, JIRA, and Linear apps) and uses code walking via Abstract Syntax Trees (ASTs) combined with LLMs to analyze projects. From there, engineers can:

 

Catch bugs in pull requests

 

Summarize code changes

 

Research their codebase via natural language queries

 

Meanwhile, product leaders can get real-time visibility into engineering output, team priorities, and productivity — without relying on endless meetings or status updates.

 

Macroscope positions itself in the competitive code review and developer tooling space alongside players like CodeRabbit, Cursor Bugbot, Graphite Diamond, and Greptile. In internal benchmarks, the company claims it caught 5% more bugs than the next-best tool while reducing noise by generating 75% fewer comments.

 

The product is priced at $30 per developer per month (with a five-seat minimum) and already counts startups and enterprises like XMTP, United Masters, Bilt, Class.com, and ParkHub among early customers.

 

Based in San Francisco, Macroscope has raised $40 million to date, including a $30 million Series A led by Lightspeed’s Michael Mignano, with backing from Thrive Capital, GV, and Adverb. Its 20-person team is now focused on scaling adoption.