outbound depends on
Domains, mailboxes, DNS and warmup are not exciting. They are also the difference between a campaign that lands and one that gets filtered before a human ever sees it. That is the whole company.
Eight people running infrastructure
for 3,400+ outbound teams
We kept rebuilding the same spreadsheet
Before Infrabox, the three of us ran outbound infrastructure for a handful of agencies out of a shared office in Delhi. Every new client meant the same week of work: buy a batch of domains, provision mailboxes across two or three providers, hand-write SPF and DKIM records, wait for propagation, then babysit warmup in a shared spreadsheet that someone inevitably broke by sorting the wrong column.
We built internal scripts to speed it up, then better scripts, then a small dashboard just for our own team. By late 2022 we had rebuilt the same tool four times for four different contracts: DNS orchestration, mailbox provisioning, a warmup scheduler, and a health checker duct-taped together with cron jobs.
In 2023 we stopped rebuilding it privately and started shipping it as a product. Infrabox is that internal tool, hardened, generalized, and now running the sending fleets of thousands of teams instead of just ours.
Last edited by Priya, 11:47pm: “fixed?? check again”
Auto-verified every 15 minutes, no manual checks
Opinions, not platitudes
Five years of running sending infrastructure taught us a few things we hold to on every roadmap call.
Infrastructure should be boring
If a domain, a mailbox or a DNS record ever needs your attention twice, we consider that a bug in Infrabox, not a fact of life.
You own the assets, not us
Domains, mailboxes and reputation are registered and billed in your name. Leave any time and take every asset with you.
Deliverability is measured, not promised
We publish the placement, bounce and blocklist data our own detection sees. No vague ‘inbox guarantee’ language anywhere.
Ship the unglamorous thing
A correct MTA-STS record ships before a nicer settings page does. The roadmap is ordered by what breaks sends, not what demos well.
Support from people who provisioned the fleet
Everyone on the deliverability rota has personally set up a cold domain. Tickets get answered with commands, not scripts.
Process that matches the product
The same discipline we ask of a sending fleet, we ask of ourselves.
Remote-first, always
No headquarters. The team works across 9 time zones and meets in person twice a year.
4-day release cadence
Every change to the console or API ships on a fixed four-day train: small diffs, fast rollback.
On-call deliverability rota
A named engineer owns placement and blocklist alerts every week, with a 15-minute response target.
Public changelog
Every shipped change is logged with what changed and why, including the ones that were reverted.
2023 to now
A short company, a long list of unglamorous milestones.
- 2023
The scripts become a product
Three of us in Delhi stop rebuilding DNS + provisioning tooling per client and start building it once, properly, as Infrabox.
- 2023
First 100 domains provisioned
Early access opens to five agency teams we used to consult for. All 100 domains are still sending today.
- 2024
Automated warmup network launches
A private peer network of warm mailboxes replaces the manual spreadsheet rotation entirely.
- 2024
Reputation Radar ships
Continuous blocklist and placement monitoring across every sender in a workspace, not just spot checks.
- 2025
10,000th domain onboarded
Fleet crosses 10,000 managed domains. Median time to first send drops to 6 minutes.
- 2025
Developer API goes public
Provisioning moves from the console into CI pipelines. Agencies script their entire fleet setup.
- 2026
46 countries, one uptime number
Sending infrastructure now spans 46 countries with a single 99.98% platform uptime figure across all of it.
The fleet, at a glance
Three years in, the boring layer is running at a scale we did not plan for.
0.0M+
Mailboxes provisioned
0K+
Domains under management
0
Countries with active senders
0.00%
Platform uptime, trailing year
Eight people, one sending fleet
Small on purpose. Everyone here has personally provisioned, warmed, or rescued a cold domain.
Priya Natarajan
Co-founder & CEO
Ran deliverability for three outbound agencies before Infrabox existed.
Marcus Feldt
Co-founder & CTO
Wrote the first DNS orchestration script Infrabox is still built on top of.
Aiko Tanaka
Head of Deliverability
Owns the warmup network and the blocklist-monitoring rota.
Diego Suárez
Staff Engineer, Provisioning
Built the bulk mailbox provisioning path that ships in under a minute.
Ngozi Adeyemi
Product Design Lead
Designs the console so a fleet of 500 mailboxes still reads at a glance.
Felix Bergström
Engineering Manager, Platform
Keeps the API and the console on the same release train, every four days.
Sana Malik
Customer Deliverability Engineer
First responder on the on-call rota when a domain's placement drops.
Owen Kowalski
Head of Revenue
Talks to agencies and RevOps teams about fleets larger than 10,000 mailboxes.
Open roles
Join a small team that owns the unglamorous layer end to end, and gets paged for it.
Run your sending fleet on infrastructure you own
Start provisioning in minutes, or talk to the team that built the fleet you'd be joining.