Location: Islamabad, Pakistan
Employment Type: Full-time
About the role
We're a bootstrapped GenAI platform serving image and video generation at real scale — Fortune 500 clients, $35M+ ARR, and traffic that doesn't wait for you to catch up. We need a backend engineer who thinks in systems, not just services. You'll own the architecture behind our generation pipelines, inference orchestration, and the infrastructure that keeps latency low while volume climbs.
This is a hands-on architect role. You will design and write code, not just draw diagrams.
What you'll do
Design and own backend systems that handle high-throughput, asynchronous workloads (generation queues, worker orchestration, media processing)
Architect for scale and reliability — you'll make the calls on data flow, service boundaries, and failure handling
Build and optimize async pipelines: job queuing, retries, backpressure, dead-letter handling, and everything that keeps millions of requests moving
Improve performance and cost at the infrastructure level — inference is expensive, and efficiency is a product feature here
Set technical direction and raise the engineering bar for the team through review and mentorship
What we're looking for
Strong backend experience (5+ years) with real ownership of production systems at scale
Proven system design skills — you can defend your architecture decisions with tradeoffs, not buzzwords
Hands-on experience with asynchronous pipelines, message queues (Kafka, RabbitMQ, SQS, or similar), and distributed workers
Comfort with high-concurrency, high-throughput environments
Ability to work independently in a fast, low-process, bootstrapped environment
Nice to have
Experience with Go (or willingness to work in it — we lean on Go for backend)
Exposure to ML/AI infrastructure, GPU workloads, or media processing pipelines
Track record contributing to open source (we look at GitHub signal)
Why ImagineArt
Real revenue, real users, no VC treadmill — decisions are made by engineers and founders, fast
Genuinely hard technical problems at the intersection of AI and scale
Fully remote, full-time