Schedule A Call

Providing enterprise IT leaders independent, evidence-based visibility into the performance of their technology vendors.

Knit.ai is a vendor performance management platform that gives enterprise IT leaders independent, evidence-based visibility into how their technology vendors are performing against SLAs — transforming support ticket data into accountability tools for QBRs, escalations, and contract renewals.
Work With Us

{ How we made it happen }

Problem

  • Knit.ai was built to solve a specific, well-documented gap in enterprise IT operations: the moment a support ticket leaves an internal ITSM system and goes to a vendor portal, the enterprise loses visibility. Traditional tools like ServiceNow manage internal workflows well, but once a case is handed off to a vendor, the vendor owns the data, the metrics, and the narrative. IT leaders negotiate contract renewals, defend project delays, and manage vendor relationships without any independent record of whether SLA commitments were honored or response timelines were met. PJ Kirner — a technology executive with deep roots in enterprise security and infrastructure — founded Knit.ai to build the missing data layer. The platform ingests vendor support ticket activity directly from email receipts, classifies cases using machine learning and natural language processing, and surfaces independent performance evidence that IT teams can bring into QBRs, escalations, and renewals. The engineering challenge was twofold: a production-grade frontend that enterprise buyers would trust, and a data pipeline capable of ingesting, classifying, and surfacing vendor support activity at enterprise scale. PJ needed engineers who could work autonomously inside a lean startup — contributors who would treat the codebase as their own, not contractors waiting for tickets. He was also notably careful about who he added to the team; every seat required demonstrated trust, not just available capacity.

Approach

  • Admios began the engagement with a single senior fullstack engineer embedded across the full product surface. The initial focus was the Knit.ai frontend — a Next.js App Router application with TypeScript, TanStack Query, and Material UI, compiling to static files served by NGINX in a Docker container. In parallel, the engineer built out a Python end-to-end test suite using Puppeteer to validate OAuth2 authentication flows across all supported identity providers — ensuring each release didn't break the sign-in paths enterprise customers depend on. As the platform's data architecture matured, the scope expanded to include cloud infrastructure work, including Google Cloud Storage configuration via Pulumi to support the upload and data intake pipeline. As the engagement deepened and PJ's confidence in the Admios team grew, the team expanded. A second frontend engineer joined approximately one year in, taking on feature development, component work, and pull request reviews — introducing paired code review as a quality gate before any feature reached staging. A third engineer joined in late 2025, focused on the data intake pipeline: specifically the ticket loader responsible for ingesting and classifying the vendor support data that powers Knit.ai's performance reporting. The team operates within Knit.ai's sprint cadence using Linear for issue tracking and GitHub for version control. Every expansion of the Admios team was earned through performance — PJ sets a high bar for who he adds, and the account has grown because the work has consistently met it.

Results

  • Knit.ai has moved from a development-phase product to a live platform with paying enterprise customers — a milestone that validates the commercial potential of the product PJ set out to build and raises the operational stakes of every release the Admios team ships. The Admios engineers have contributed across every layer of what Knit.ai sells: the frontend that enterprise buyers see and trust, the test infrastructure that protects production authentication flows, and the data intake pipeline that powers the performance evidence at the core of the platform's value proposition. The growth of the Admios team on this account tells its own story. The decision to expand from one engineer to three came directly from PJ's observation that the engineers already in place were contributing at a level that justified adding more capacity. That's a meaningful signal from a founder who is the primary judge of engineering quality and who is known for being selective about who joins his team. Growth here was not sold — it was earned. Admios remains an active partner as Knit.ai continues building out the signal library, refining the data intake pipeline, and scaling toward the production robustness that enterprise buyers require. It is one of Admios's longest-running active engagements, and it reflects the kind of technically demanding, high-trust work that Admios is built to support over the long term.

{ Overview }

Industry
Enterprise Software / IT Operations
Company Size
Seed-stage startup
Dev Team Size
3 engineers (current)

{ Key Results }

Team growth driven by performance. The Knit.ai account has grown from one Admios engineer to three, with each expansion driven by confidence in what the team was delivering.

Platform live with paying customers. The product crossed from development into production with real, paying enterprise customers, a milestone that validates the platform's market fit and raises the stakes of every release the Admios team is responsible for shipping.

Full-stack coverage across all product layers. The Admios team spans the frontend application, Python end-to-end test infrastructure, cloud storage configuration, and the data intake pipeline — a breadth of contribution that reflects how deeply embedded the team is in the product, not just a single workstream.

{ Stack Used }

No items found.