Origin Story
The insight from the 11pm dashboard
Priya Raman spent years building and managing support operations for SaaS teams. The pattern she saw repeatedly was the same: a queue of 60-100 tickets waiting every morning, most of them refund requests and order status checks that agents would spend the first two hours resolving before touching anything that required judgment.
The classification tools existed. Chatbots could deflect. But nothing actually closed the ticket, which meant agents still opened every morning to the same backlog and support velocity tracked to headcount instead of automation rate.
Replylume was built to solve that specific gap: not to classify, not to deflect, but to close tier-1 tickets end to end by running the full action sequence that agents were spending their first two hours doing manually every day.
"About 70% of the tier-1 queue is lookup-and-act, not judgment. Someone placed an order, something went wrong, they need the record looked up and the refund issued. That is a four-step process that takes a system 8 seconds and a human 4 minutes. The math was obvious once I named it."
Team
Three people, one focus
Priya Raman
CEO & Co-Founder
Spent years building support operations for SaaS teams, with a focus on how support velocity ties to retention and churn prevention. Watched tier-1 queue buildup become a recurring operational bottleneck before deciding to build the resolution layer directly.
Daniel O'Reilly
CTO & Co-Founder
Built API orchestration layers connecting SaaS toolchains across support, billing, and CRM platforms. Focused on the reliability and auditability of automated action sequences: systems that act need logs you can trust and fail-open escalation you can trace.
Amara Osei
Head of Customer Success
Experienced support operations practitioner who has managed tier-1 teams at SaaS companies. Understands the buyer's day-to-day from the inside: what makes automation trustworthy versus frustrating, and where the handoff to a human agent has to work without fail.
Values
How we make decisions
Specificity
We solve one problem: tier-1 ticket resolution. We will not expand into general support automation, AI helpdesk, or chat platform territory. The teams that trust Replylume with their billing data deserve a tool that does exactly one thing and does it correctly.
Honesty
We show the automation rate from our actual pilot cohort, not a headline number. We show our pricing without a "contact for pricing" button. We do not claim certifications we have not completed. If Replylume is not the right fit for your support stack, we will say so in the demo.
Trust
Replylume acts on billing data on your behalf. That access requires an audit trail your security team can read, scope limits that do not creep, and an escalation path that works without fail. We log everything because automated systems that act invisibly are systems that erode trust over time.