Founding Engineer - AI Infra & Scale

We're hiring a Founding Engineer focused on Scale, Stability and AI infra. As we ramp from thousands to millions of AI and enrichment calls daily, latency, throughput, and reliability become first-class product features. This role owns that frontier, making sure BitScale stays fast, predictable, and unbreakable as load grows 10x and then 10x again. You'll go deep into our Go backend, our AI/LLM infra, our data layer, and the other performance edges wherever scale hurts the most.
Responsibilities:
Own performance and reliability of our Go backend services from API latency to background workers to long-running enrichment pipelines.
Design and build rate limiting, queueing, and backpressure systems that keep the platform predictable under spiky, bursty load.
Scale our NoSQL and search layer schema design, query optimisation, sharding, hot-key mitigation, and Elasticsearch / OpenSearch tuning.
Optimise our AI/LLM infrastructure gateways, token routing, batching, caching, and fallbacks across model providers so we ship faster responses at lower cost.
Hunt down latency on both frontend and backend perf edges where it matters: initial paint, hydration, data fetching patterns, real-time updates.
Build the observability layer: metrics, traces, alerts, runbooks so we catch regressions before customers do.
Lead load testing, capacity planning, and incident response as we step up to each new order of magnitude.
Directly influence architecture and roadmap with the founders.
Requirements:
2-3+ years of experience building production backend systems that have actually been stressed by real traffic.
Someone who treats p99 latency, throughput, and error rates as product metrics, not afterthoughts.
Deep ownership mindset: you don't just file a fix, you make sure it never happens again.
Comfortable in a high-intensity, highly agile environment.
Pragmatic engineer: knows when to ship scrappy and when to go deep on architecture.
Hungry to learn new languages, tools, and stacks on their own.
Pre-requisite Tech Stack: Go, NoSQL (MongoDB / DynamoDB / similar), Redis, Elasticsearch / OpenSearch, AWS.
Good to Have:
Previous start-up or early-stage experience.
Deep interest in system design, performance engineering, and scaling.
Experience with event-driven systems, queues, or workflow engines.
Comfort going below the abstraction: profilers, flame graphs, EXPLAIN plans, network traces.
Success in your first 6 months:
p99 latency cut by 50%+ on the hottest paths in the product.
Platform handles 10x current load without rate-limit-driven incidents or outages.
LLM cost and latency measurably down through smarter routing, batching, and caching.
A clear, instrumented view of where the next bottleneck is before it bites.

Experience
3-6 yrs