Introducing Macroscope

After a year of building quietly, we’re excited to announce Macroscope: an AI-powered understanding engine that gives leaders clarity while saving engineers time.

The problem

As companies are shipping exponentially more code, humans are reviewing far less of it.

This makes an old problem worse: it’s incredibly hard to see what’s actually happening and to answer questions like:

• What did we get done this week?
• What progress are we making against priorities?
• How did we evolve the product this week?

Every founder, engineering manager, and product leader asks these questions constantly. Getting clear, accurate answers is far harder than it should be.

Most companies resort to a soul-crushing mix of endless meetings and scattered tools (JIRA tickets, spreadsheets, etc). Engineers waste hours on “the work around the work”—writing updates instead of building. Even then, the results are often stale, diluted, sugar-coated proxies for the truth.

The result: bureaucratic process hell—draining, demoralizing, slow. It’s the kind of big-company bullshit that drives great engineers to faster-moving startups. We know, because we’ve lived it.

Macroscope fixes this

Macroscope is an AI system that leverages your codebase (and tools like Linear and JIRA) to synthesize, summarize, and answer what’s happening.

Macroscope dashboard

For leaders, Macroscope delivers:

• Real-time summaries of product development, from granular commit feeds to weekly digests.
• Productivity insights showing trends in team output (including AI agents).
• Ground-truth engineering allocation revealing what teams actually prioritize—not just what was planned.
• On-demand answers to natural-language questions about the product, code, or dev activity—without distracting engineers.

For engineers, Macroscope saves time by:

• Finding and suggesting fixes for bugs in PRs.
• Writing succinct and accurate PR summaries automatically.
• Delivering a searchable and summarized feed of how the codebase is changing.
• Answering code-research questions—from debugging to onboarding to exploring poorly documented systems.

Macroscope Code Comment

How it works

Underpinning Macroscope is a perception layer that processes and synthesizes activity from your product development process.

The most important part of this perception layer is our “code walking” system. Walkers traverse the Abstract Syntax Tree (AST) of your code, constructing a graph of your entire codebase. For any given piece of code, Macroscope gathers the most relevant code references to help state-of-the-art language models understand critical nuances of how that code impacts your codebase and product.

This approach allows Macroscope to:

• Understand even the most complex, multi-repository codebases.
• Identify bugs in your Pull Requests that other code review tools miss
• Generate coherent, accurate technical summaries that reflect the true state of your product.

In our experience, this technique plays a large role in avoiding hallucinations, mischaracterizations and LLM gibberish—often the result of missing nuance and context—versus higher quality, lower noise output that makes customers say: “Wow. Macroscope gets it.”

Macroscope Diagram

Impact

Macroscope is already driving impact with many customers, from fast-moving AI startups to scaled multi-billion dollar enterprises with hundreds of engineers.

Here’s what some of our customers have said:

Kris Stokking, SVP Product Development, Class.com: “Macroscope provides us with invaluable ground truth about our evolving product, giving clear visibility into rapidly changing and complex components. It’s become a critical time saver for product development—providing useful feedback to developers during code reviews, reducing time-to-root-cause in troubleshooting, helping organize project plans and release artifacts, and illuminating areas that have lacked clear documentation.”

Logan Fisher, CTO, ParkHub: “The time we’ve saved and the insights we’ve gained have changed the way we work. From creating custom focus areas that give us sharper visibility into what really matters, to automating product and technical release notes, Macroscope just gets it. They’ve cracked the code on turning huge amounts of complex data into clear, actionable insights—and they’ve made it effortless for us to make smarter decisions, faster.”

Jonathan Freeland, VP of Engineering, Seed.com “Macroscope has become essential for maintaining visibility across our engineering organization at Seed. From automated PR summaries and reviews to weekly updates, it gives my team and I the ground truth on what's actually happening in our codebase and what we are accomplishing. Our engineers get to spend more time doing what they do best: shipping great code.”

Scott Belsky, Cofounder of Behance @ Founder A24 Labs: “Macroscope has become a core part of our engineering team, bringing some new superpowers in productivity and keeping us all aligned and up to speed on what's getting done every day.”

Marcel Molina, CTO & Cofounder, Particle: “Macroscope is like having a distinguished engineer tech lead who's read every diff, understands every project, and can answer any question about your codebase instantly. It allows us to focus on shipping instead of process.”

Nick Molnar, CTO @ XMTP: "We've used just about every AI-driven PR assistant out there: the signal to noise from Macroscope is the best I've seen. The PR descriptions are better than what we would have written by hand, and when it flags an issue it's almost always a real bug.”

Code Review Performance

TLDR: According to our internal benchmark of 100+ real-world bugs, Macroscope caught 5% more bugs from the dataset and generated 75% fewer comments compared to the second-best tool.

In order to evaluate Macroscope’s code review, we built a dataset of 100+ real-world bugs from 45 popular open-source repositories across the 8 programming languages we support. This dataset only includes runtime bugs, and excludes subjective styling or quality issues.

We benchmarked Macroscope alongside several code review tools (CodeRabbit, Cursor Bugbot, Graphite Diamond, Greptile) using this dataset, focusing on three key dimensions:

• Detection rate of bugs from our dataset
• Comment volume
• Price

The results of our benchmark indicate Macroscope achieved the highest detection rate of bugs from our dataset. CodeRabbit and Cursor Bugbot were close behind, followed by Greptile and Graphite Diamond.

To evaluate the volume of comments generated by each tool (i.e. how “loud” or “quiet” a tool is), we measured the average number of comments left on each Pull Request we opened. We did not assess the quality, value or correctness of all of these comments. We simply quantified the average number of review comments left per opened PR, so that we could illustrate a normalized volume meter for how “loud” each tool was on its default settings. Macroscope’s volume ranked exactly in the middle of the five tools we evaluated, while CodeRabbit was significantly the “loudest” and Graphite Diamond was the “quietest”.

For more detail on our methodology and benchmarking results, see our detailed blog post here. We plan to continuously rerun internal benchmarks to make sure we’re delivering the best code review experience for our customers. Periodically, we’ll publish the results of these internal benchmarks to keep the broader community updated on our progress.

Just the beginning

We raised a $30M Series A led by Lightspeed, with participation from Adverb, Thrive Capital and Google Ventures, bringing our total funding to $40M. We’re grateful to have the backing of some of the best investors in the world who share our conviction for the future of software development. Looking forward, Macroscope will weave together a richer view of how work gets done across your company—and evolve from understanding to directly helping build your product. Today’s launch is just the first step.

Macroscope was founded by repeat entrepreneurs Kayvon Beykpour, Joe Bernstein and Rob Bishop. Kayvon and Joe co-founded Periscope, the live streaming app that was acquired by Twitter in 2015– where Kayvon went on to lead Consumer product and engineering until 2022. Rob co-founded Magic Pony, an ML/ computer vision startup that was acquired by Twitter in 2016, and was the first employee at Raspberry Pi. We’re joined by an incredible team of founding engineers (Aaron Wasserman and Jonathan Le), and around a dozen other talented teammates passionate about building the future of software development. We’d love to hear from you if you’re interested in joining us.

Macroscope team