Be kind to each other. Leading engineering at @Macroscope. Previously led tech @unitedmasters and built things like @TwitterSpaces and @PeriscopeCo.
SF Bay Area
Joined November 2008
- Tweets788
- Following241
- Followers1.3K
- Likes1.5K
Pablo Jablonski retweeted
approvability is macroscope's agent that auto-approves PRs that are safe to merge.
approvability now runs on an upgraded model, so it handles longer, more complex policies more reliably. it's also more permissive by default, so it approves a larger set of PRs that are safe to merge.
there's also a new "Mode" setting (settings → repos → approvability). Permissive (the new default) favors automatic approval when supported by verifiable evidence that the PR is safe, even if the code touches more sensitive areas. Strict is closer to the previous behavior, and requires human review when meaningful risk or uncertainty remains after considering safeguards.
custom policy guidelines you've configured still take priority in both modes.
Pablo Jablonski retweeted
There are some new top models leading our code review benchmark.
🥇 Claude Opus 5.5 — 82.3 (f1 score)
🥈 GPT-6 Sol — 80.3
🥉 GPT-6 Astra — 78.0
See full results here: macroscope.com/benchmark
Pablo Jablonski retweeted
Production is the worst place to find out if your code reviewer is good enough
MacroscopeBench measures it first, against real bugs that shipped, just not in your repo.
It is now the only trusted benchmark for code review on @FireworksAI_HQ SII Index fireworks.ai/specialized-int…
So much thought, time, experimentation, and love has gone into building this benchmark at Macroscope. It would be impossible to do the work we do without it, and we're excited to unveil it to the world.
Pablo Jablonski retweeted
I’m glad @strawgate's standards for coffee are lower than his standards for code review because many coffees were brewed by yours truly in the creation of this video. I was thrilled to sit down with Bill, who is the head of product at @pydantic, to hear about how he and the team use @Macroscope. It’s an absolute joy to work with Pydantic and to have a small part in supporting their work.
Very excited to share what we've been cooking ! We've been using Murmur to build at Macroscope for a few months now, and it's completely changed what we can do. Very excited to give other teams the chance to give it a shot as well
Introducing ULTRA! Our best code review pipeline ever, and bring better every day
Today we’re expanding Detection Mode for code review so that you can choose the ideal performance profile for each review:
Ultra: Frontier bug detection for the most consequential reviews. Leverages a much higher number of parallel agents using more advanced coding models. Useful when maximizing bug detection matters more than cost.
Balanced (default): Industry leading high bug detection. Recommended for teams who use Agents to review PR comments, where the inconvenience of occasional false positives is minimal.
Precise: Similar bug detection as Balanced mode, but higher confidence findings due to additional agentic validation to reduce noise from false positives and low-value findings.
Budget: Fast, low cost bug detection.
We’re all accustomed to choosing different intelligence-level models for different tasks… the same should be true for code review.
I’m particularly proud of this launch because Macroscope’s Balanced pipeline is already exceptional and (according to the latest independent benchmarks from Martian), has the highest f1 score of any other code reviewer. Our new Ultra pipeline is a further 10% improvement on this!
Pablo Jablonski retweeted
We just launched a new version of @Macroscope code review that finds significantly more bugs, including 71% more critical issues, 27% more high-severity issues, 50% more medium-severity issues, and 31% more valid issues overall in our benchmarks, at the expense of a modest increase in false positives and review latency.
I’ve been lucky enough to be a beta user of Casa for a long time now, and it has already become essential to my life. The team is incredible and the app makes so many things that are usually a pain as a homeowner incredibly easy to manage. Huge congrats to the team, and if you own a home, you HAVE to check it out!
After two years of building under wraps, today we're announcing Casa – your personal property manager. We've raised $27M to redefine the homeownership experience from the ground up.
We believe your home is your most treasured asset, emotionally & financially. It shouldn't also be a second job.
Most homeowners are on their own – expected to have the time, expertise, and relationships to keep things running. Finding a plumber you can trust. Remembering when the HVAC was last serviced. Knowing what's actually wrong before someone shows up to fix it.
Casa gives every homeowner what used to be reserved for the few: a dedicated team that knows your home deeply, handles the work, and stays in your corner. We're enabling this by building a deep, technical understanding of every home we serve – something that's never existed before, across 100 million single-family homes in the country.
For $199/mo, membership includes:
- A complete inventory of your home, built using specialized hardware & software
- 1.5 hours of handyman time every month (and it rolls over)
- Unlimited Concierge requests to take on virtually any home project
- Custom, proactive care plans built specifically for your home
- Weekly package and donation pickups
- Scheduling and payments for your regular vendors
- Utility and property tax monitoring
…and we’re just getting started, with more benefits on the way to make the experience of owning your home as magical as it always should have been.
Available now in the SF Bay Area and Los Angeles. Reserve your spot everywhere else.
→ getcasa.com
Pablo Jablonski retweeted
Check Run Agents, now available in @Macroscope. we made a video to show you how it works!
these are super flexible agents you control that let you enforce your team's coding conventions, workflows, and compliance checks during code review. just define in markdown and include trigger rules, model choice/reasoning levels, and tool access (including any 3rd party integrations and MCP servers you've already connected to macroscope)-- then the agent spawns as a Check in GitHub.
we've been delighted by the creativity and complexity of what our customers have been doing with check run agents: enforcing style conventions, coordinating multistep db migration workflows, ensuring public-facing docs are updated in sync with code changes, and more.
Pablo Jablonski retweeted
We 🚢 a TON of new features this week ↓
→ Ask Macroscope to create PRs from Slack
→ Webhooks to invoke Macroscope from anywhere
→ Code Review validates your PRs against linked @Jira or @linear issues
→ New native integrations w/ BigQuery, @LaunchDarkly, and @PostHog
→ Code Review: "AI Prompt" for easy copy/pasting to your coding agent
...and more
Pablo Jablonski retweeted
Every company is drowning in pull requests. It's no longer tenable for humans to review every single PR.
@macroscope is fixing this. Today we launched Approvability in beta: it looks at the blast radius and security risk of your PRs, along with any custom policy guardrails you define, then auto-approves the PRs that don’t need human attention.
Now available in beta to all customers.
Take a peak behind the curtain at the incredible tech that powers Macroscope's code review
Both Anthropic and OpenAI dropped new models today. For some teams, that means hours/weeks/days of re-tuning prompts. For us, it means hitting "run." We wrote about why we stopped writing prompts: macroscope.com/blog/we-stopp…
Not incremental, step-change.
Our AI code review now finds more bugs than human reviewers. Our precision (above 95%!!) means no more AI slop that annoy you.
This is an inflection point in making AI code review truly indespensable, and we're just getting started 🚀
Today we're releasing Macroscope Code Review v3. Based on our internal benchmarks:
→ Detects up to 3.5x more bugs that would cause real production damage (data loss, security breaches, crashes, etc). The kind you'd block a PR over.
→ Precision increased to 98% (up from 75%) which means significantly fewer false positives.
→ Leaves 22% fewer comments overall, including 64% fewer nitpicks in Python and 80% fewer in TypeScript.
Pablo Jablonski retweeted
Delightful demo moment that would have been inconceivable a year or two ago: While doing an onboarding yesterday, I noticed a bug in our web app. Mid-meeting I posted the bug in Slack flagging one of our devs and asking @Macroscope to make a linear ticket while + investigate and fix in parallel.
15min later Macroscope had made the ticket, found the issue, fixed it in a PR and validated no bugs in code review. End to end the fix was merged, tested (via live vercel preview link) and deployed to prod within 20min-- before the meeting was even over. Inadvertently turning a lemon into lemonade and a made for a great demo.
What a time ...
It's been awesome to help the GoodDay team be more productive and build great products. A great example of what we built Macroscope to do!
GoodDay's engineering team started shipping faster with AI coding tools. But @gooddaysoftware Co-Founder/CPO @dave_wardell couldn't keep up and engineers were spending hours answering his questions about what changed in the codebase.
With Macroscope, the whole team, including Product and Customer Success, now gets instant answers and automated reporting. Painful weekly overhead eliminated.
Read the case study ➡️ blog.macroscope.com/blog/cas…
Pablo Jablonski retweeted
for anybody motivated to try an alternative AI code review tool, we're happy to offer you 3 months free of @Macroscope
use coupon code XMASCOPE during Stripe checkout, valid until sunday EOD.
Aiden - you clearly have no idea about what you are talking about. We have more users than everyone you just mentioned (combined). See aitooltracker.dev/
Still, I am curious how we can make it simpler and fool proof for indie developers. Maybe simpler controls and a lot of handholding can help.
