AI infrastructure for the AI marketing platform.
Model serving, evaluation, and the attribution loop underneath Pixis's AI marketing platform — with the infrastructure discipline to sustain a fast-growth SaaS trajectory.
Client
Pixis
Sector
Technology & SaaS
Duration
10 months
Team
7 (AI engineers, an SRE embed, a platform architect, data engineers)
Where Pixis was when we started.
Pixis is an AI-native marketing platform serving growth teams at consumer brands. As the company scaled from Series A to Series C, the internal AI infrastructure needed to move from 'a few models in a notebook' to production infrastructure with the evaluation, cost discipline, and serving profile a SaaS company requires.
The problem, unvarnished.
- Model serving cost was scaling faster than revenue as customers grew
- Evaluation was manual — every new model release depended on human review by the DS team
- Attribution model outputs varied between customers in ways the team couldn't debug at speed
- The AI programme was operating without the SRE and cost-discipline patterns SaaS scale demands
How we scoped and sequenced the work.
01
Cost-aware model serving
Migrated model serving to a tiered infrastructure — hot path on GPU, warm path on quantised CPU, cold path on batch — with routing based on latency budget and customer plan. Serving cost per query fell without touching the product.
02
Evaluation as continuous product surface
Built the eval platform: golden datasets per model family, continuous eval on every deploy, regression alerts on drift. Every new model has to pass its eval to ship. The DS team stopped being the release blocker.
03
Attribution consistency across customers
Standardised the attribution model surface across customer segments. Where per-customer variation was necessary, it was made explicit and auditable — not a mystery in the config.
04
SRE embed for AI infrastructure
Embedded an SRE with the AI platform team for six months. On-call practices, incident review, capacity planning, and cost dashboards — the boring operational discipline that AI infra needs at scale.
What we shipped.
A production AI infrastructure stack underneath Pixis's marketing platform, with tiered serving, continuous evaluation, and the SRE discipline to sustain SaaS growth. Attribution model surface standardised across customer segments.
The numbers that matter.
62%
Reduction in serving cost per query
Achieved through tiered serving and model quantisation without measurable degradation in customer-facing latency.
0
Human release-review gates
Continuous evaluation replaced manual DS review as the release gate for model updates.
14
Attribution channels standardised
Consistent attribution surface across customer segments with explicit, auditable per-customer variation where required.
What we built it on.
Bring us your technology & saas brief.
Book a 30-minute discovery call. A senior practitioner from the same practice that shipped this engagement will scope yours.
or email contact@headify.com