Pinned Tweet
Meet the superintelligent AI SRE that starts investigating before you do. Traversal Workers are now generally available.
1
3
13
974
We’re excited to welcome @nubank as a Traversal customer.
Nubank is the largest digital bank in Latin America, serving 139 million customers across Brazil, Mexico, and Colombia. As Nubank rapidly expands its customer base, products, and markets, its engineering organization is operating thousands of services where even brief production issues can affect millions of customers.
@velez_david, Founder and Global CEO of Nubank, shared: "Five weeks into our Traversal rollout, and it’s already becoming part of how our on-call engineers work incidents. We're expanding it wave by wave as those results hold up. It's early, but it's the kind of leverage we want AI to give our engineering teams."
Thank you to the Nubank team for trusting Traversal. We’re excited for what’s ahead.
Nubank is the largest digital bank in Latin America, serving 139 million customers across Brazil, Mexico, and Colombia. As Nubank rapidly expands its customer base, products, and markets, its engineering organization is operating thousands of services where even brief production issues can affect millions of customers.
@velez_david, Founder and Global CEO of Nubank, shared: "Five weeks into our Traversal rollout, and it’s already becoming part of how our on-call engineers work incidents. We're expanding it wave by wave as those results hold up. It's early, but it's the kind of leverage we want AI to give our engineering teams."
Thank you to the Nubank team for trusting Traversal. We’re excited for what’s ahead.
0
1
5
186
How do you keep an AI agent reliable while investigating broken production systems?
Our latest engineering blog breaks down how Traversal’s agent runtime was rebuilt on @temporalio for granular retries, resumable investigations, elastic capacity, and mid-run steering.
Read GTvysB
Our latest engineering blog breaks down how Traversal’s agent runtime was rebuilt on @temporalio for granular retries, resumable investigations, elastic capacity, and mid-run steering.
Read GTvysB
0
1
7
751
Tuesday night at re:Invent, we’re taking over Gymkhana. All of it.
Traversal and Factory are hosting The Agents Table, an invite-only dinner bringing together senior engineering and technology leaders over Michelin-caliber Indian food and drinks.
The evening will kick off with ste, CTO of American Express, followed by a conversation moderated by @BogieBalkansky of Sequoia Capital with five CEOs building across AI and infrastructure:
@_anish_agarwal - CEO, Traversal
@matanSF - CEO, Factory
@hwchase17 - CEO, LangChain
@SamarAtTemporal - CEO, Temporal
@paraga - CEO, Parallel Web Systems
December 1. Gymkhana at ARIA. The Agents Table.
Request to attend: luma.com/agents-table-a…
Traversal and Factory are hosting The Agents Table, an invite-only dinner bringing together senior engineering and technology leaders over Michelin-caliber Indian food and drinks.
The evening will kick off with ste, CTO of American Express, followed by a conversation moderated by @BogieBalkansky of Sequoia Capital with five CEOs building across AI and infrastructure:
@_anish_agarwal - CEO, Traversal
@matanSF - CEO, Factory
@hwchase17 - CEO, LangChain
@SamarAtTemporal - CEO, Temporal
@paraga - CEO, Parallel Web Systems
December 1. Gymkhana at ARIA. The Agents Table.
Request to attend: luma.com/agents-table-a…
1
2
13
454
Traversal ranked in the Top 50 on Paraform’s Talent Density Index, alongside teams like Anthropic, OpenAI, LangChain, and Perplexity.
From the beginning, we’ve been intentional about bringing together exceptional people from different disciplines – AI researchers, distributed ers, infrastructure experts, experienced SREs, and GTM leaders – and giving them room to take big swings at hard technical problems.
We’re proud of the team we’ve built, and we’re adding to it ... fast. Thanks to @paraformtalent for the recognition!
Explore our open roles here: traversal.com/careers?utm_so…
From the beginning, we’ve been intentional about bringing together exceptional people from different disciplines – AI researchers, distributed ers, infrastructure experts, experienced SREs, and GTM leaders – and giving them room to take big swings at hard technical problems.
We’re proud of the team we’ve built, and we’re adding to it ... fast. Thanks to @paraformtalent for the recognition!
Explore our open roles here: traversal.com/careers?utm_so…
0
0
4
179
Traversal was named a @FastCompany 2026 Next Big Things in Tech honoree for our work on Traversal Workers.
Workers don’t wait to be tagged or prompted. They proactively investigate alongside human teams and know when to contribute.
Since launching this summer, they’ve handled nts at some of the world’s largest enterprises – helping close them faster, with fewer engineers pulled into the response.
Learn more: traversal.com/blog/traversal…
#FCTechAwards
Workers don’t wait to be tagged or prompted. They proactively investigate alongside human teams and know when to contribute.
Since launching this summer, they’ve handled nts at some of the world’s largest enterprises – helping close them faster, with fewer engineers pulled into the response.
Learn more: traversal.com/blog/traversal…
#FCTechAwards
0
0
3
157
We’re excited to welcome Harvey as a Traversal customer.
More than 200,000 lawyers across 3,000+ organizations in 70+ countries use Harvey, including 80% of the Am Law 100 and five of the Fortune 10.
Harvey has also built sophisticated AI tooling for its own engineering and aking the bar for an external AI SRE particularly high.
@sgurumur , CTO of Harvey:
“We’re taking a modern approach to software development, and we build much of the technology ourselves. When we work with an external partner, the bar is high, and Traversal clears it. It helps our engineers extend the AI-native development lifecycle all the way to production as Harvey scales.”
Thank you to the @harvey team for trusting Traversal. We’re excited for what’s ahead.
More than 200,000 lawyers across 3,000+ organizations in 70+ countries use Harvey, including 80% of the Am Law 100 and five of the Fortune 10.
Harvey has also built sophisticated AI tooling for its own engineering and aking the bar for an external AI SRE particularly high.
@sgurumur , CTO of Harvey:
“We’re taking a modern approach to software development, and we build much of the technology ourselves. When we work with an external partner, the bar is high, and Traversal clears it. It helps our engineers extend the AI-native development lifecycle all the way to production as Harvey scales.”
Thank you to the @harvey team for trusting Traversal. We’re excited for what’s ahead.
0
0
6
133
The hardest part of AI SRE isn’t the agent. It’s everything the agent needs underneath it. Here’s how Traversal’s architecture comes together: traversal.com/blog/inside-tr…
0
0
5
212
ORCA-bench is now on @FireworksAI_HQ 's Specialized Intelligence Index, its new index for real-work benchmarks across industries. External validation of what we set out to build: a rigorous benchmark for production RCA.
The results are a reality check. Across 1,079 tasks, the ls hit 25.3% accuracy on medium-difficulty incidents, dropping to 10% with a vague report.
Understanding production takes much more than a frontier model.
fireworks.ai/specialized-in…
The results are a reality check. Across 1,079 tasks, the ls hit 25.3% accuracy on medium-difficulty incidents, dropping to 10% with a vague report.
Understanding production takes much more than a frontier model.
fireworks.ai/specialized-in…
0
1
12
325
Coding agents know your code. They don’t know production.
For months, customers have been using Traversal MCP to bring real production context into tools like Claude Code: topology, dependencies, recent changes, and system behavior.
Instead of just providing access to an API, ing agents a continuously updated model of production, helping flag how changes could affect their systems before they ship.
We wrote about how it works back in March: traversal.com/blog/introduci…
For months, customers have been using Traversal MCP to bring real production context into tools like Claude Code: topology, dependencies, recent changes, and system behavior.
Instead of just providing access to an API, ing agents a continuously updated model of production, helping flag how changes could affect their systems before they ship.
We wrote about how it works back in March: traversal.com/blog/introduci…
1
0
4
227
AI shouldn’t be a black box, especially in production.
That’s why Knowledge Bank™ lets you see and interact with everything Traversal knows about your environment, review what it’s learned from investigations, and add the context that only exists in your team’s heads, all in uilds that understanding automatically. You don’t have to add anything, but when there’s context you want to add, you can.
Read more here: traversal.com/blog/introduci…
That’s why Knowledge Bank™ lets you see and interact with everything Traversal knows about your environment, review what it’s learned from investigations, and add the context that only exists in your team’s heads, all in uilds that understanding automatically. You don’t have to add anything, but when there’s context you want to add, you can.
Read more here: traversal.com/blog/introduci…
0
0
5
231
Putting an AI SRE to work shouldn't be a project of its own.
Traversal Workers scan your environment and deploy themselves where they're needed – one click, no custom instructions. They join the right channels immediately, and new ones as they appear.
One console shows every vestigating.
Customers have scanned, deployed, and seen value all in the same day. It’s that easy.
More here: traversal.com/blog/traversal…
Traversal Workers scan your environment and deploy themselves where they're needed – one click, no custom instructions. They join the right channels immediately, and new ones as they appear.
One console shows every vestigating.
Customers have scanned, deployed, and seen value all in the same day. It’s that easy.
More here: traversal.com/blog/traversal…
0
0
6
291
We’re a @FastCompany Best Workplace for Innovators, recognized in AI, Automation and Machine Learning Excellence.
Building an AI SRE means solving problems that didn’t really exist a few years ago: How should production environments be represented so autonomous systems can hould AI investigate across billions of signals to find root cause? What happens when an AI agent proactively starts the investigation?
Answering those questions takes an unusual mix of AI researchers, distributed systems engineers, infrastructure experts, SREs, and GTM leaders who can turn new technical capabilities into something enterprises can actually adopt and use.
We’re hiring across the company and adding new roles every week: traversal.com/careers?utm_so…
Building an AI SRE means solving problems that didn’t really exist a few years ago: How should production environments be represented so autonomous systems can hould AI investigate across billions of signals to find root cause? What happens when an AI agent proactively starts the investigation?
Answering those questions takes an unusual mix of AI researchers, distributed systems engineers, infrastructure experts, SREs, and GTM leaders who can turn new technical capabilities into something enterprises can actually adopt and use.
We’re hiring across the company and adding new roles every week: traversal.com/careers?utm_so…
0
0
4
237
Your production context is everywhere: telemetry, code, incidents, conversations, runbooks, and plenty of tribal knowledge.
Traversal’s integrations bring all of it together for Traversal to reason across, mapping causal relationships no individual tool can see.
That connected race root cause when something breaks, or ask pretty much anything about production: check service health, investigate changes, optimize costs, and more.
No predefined schemas or heavy integration lift required.
See how Traversal can connect to your environment: traversal.com/book-a-demo?ut…
Traversal’s integrations bring all of it together for Traversal to reason across, mapping causal relationships no individual tool can see.
That connected race root cause when something breaks, or ask pretty much anything about production: check service health, investigate changes, optimize costs, and more.
No predefined schemas or heavy integration lift required.
See how Traversal can connect to your environment: traversal.com/book-a-demo?ut…
1
0
4
349
AI SRE evaluations tend to focus on accuracy and latency. Effort and time to value deserve just as much scrutiny.
How long until the system delivers value? And what does your team have to write, install, tune, migrate, and maintain to get there?
Three questions to ask when ttps://t.co/oM9uxcsTZi
How long until the system delivers value? And what does your team have to write, install, tune, migrate, and maintain to get there?
Three questions to ask when ttps://t.co/oM9uxcsTZi
0
0
2
198
100M+ entities. 10B+ causal relationships. One continuously updated model of your production environment.
Traversal’s Production World Model™ maps applications, services, infrastructure, changes, and how they connect, giving our AI SRE the real-time understanding it needs for Customers use that same foundation far beyond incidents, from service health and impact analysis to infrastructure investigation, capacity planning, and questions about production in natural language.
See what it can do in your environment: traversal.com/book-a-demo?ut…
Traversal’s Production World Model™ maps applications, services, infrastructure, changes, and how they connect, giving our AI SRE the real-time understanding it needs for Customers use that same foundation far beyond incidents, from service health and impact analysis to infrastructure investigation, capacity planning, and questions about production in natural language.
See what it can do in your environment: traversal.com/book-a-demo?ut…
0
0
3
245
A simple test for an AI SRE:
What happens when the answer isn’t in the runbook?
What happens when the answer isn’t in a markdown file?
If engineers have to document every service, dependency, and failure mode before it can investigate an incident, we’re really stretching the AI SRE should be able to figure out what nobody wrote down.
Our latest on the markdown trap: traversal.com/blog/why-an-ai…
What happens when the answer isn’t in the runbook?
What happens when the answer isn’t in a markdown file?
If engineers have to document every service, dependency, and failure mode before it can investigate an incident, we’re really stretching the AI SRE should be able to figure out what nobody wrote down.
Our latest on the markdown trap: traversal.com/blog/why-an-ai…
0
0
3
223