AI Infrastructure Engineer
The platform under your AI: gateways, GPUs, vector stores, and zero-downtime scale.
Builds and runs the substrate AI products live on, model gateways, inference infrastructure, vector databases, caching, rate limiting, and multi-provider failover.
OUTAGES YOUR CUSTOMERS NEVER SEE
WHAT THEY OWN
Concrete deliverables, not job-description poetry.
01
Model gateway & routing layer
One controlled door to every provider, keys, quotas, fallbacks, audit.
02
Inference infrastructure
Self-hosted or hybrid serving tuned for latency and unit cost.
03
Vector store operations
Indexing, sharding, and backup strategies that survive growth.
04
Caching & rate limiting
Semantic caching and throttling that cut spend without cutting quality.
05
Provider failover
Outages at OpenAI or Anthropic become a log line, not an incident.
06
Observability foundation
Latency, cost, and error budgets per feature, per customer.
PRICING
Pick the level, keep the senior oversight.
Junior
$2,400 /month
or $15/hr on Time & Material
AI-native from day one
Mid-Level
MOST HIRED$3,500 /month
or $22/hr on Time & Material
Independent feature ownership
Senior
$4,600 /month
or $29/hr on Time & Material
Architecture & judgment
Dedicated engineers are billed monthly; Time & Material is billed hourly on tracked actuals. The free trial week applies to every dedicated hire.
YOU NEED THIS ROLE IF
One provider outage takes your product down with it
AI spend is a single scary invoice nobody can decompose
Every team calls model APIs their own creative way
BY END OF WEEK ONE
Mapped every model call path in your product
Put a gateway in front of the chaos
Broke down cost per feature and per customer
Set the first latency and error budgets
OUTCOMES YOU CAN MEASURE
Provider outages your customers never see
AI unit economics per feature
Latency budgets that hold at scale
One governed path to every model
PAIRS WELL WITH