Go Developer

Go Developer Intern — AI Inference Infrastructure

Company:

Gpu.net

Location:

Bangalore - Manaya tech park

Duration:

[3 ] months, with possibility of a full-time offer

Stipend:

10000

About us

We run a large-scale AI inference cluster built on next-generation accelerator hardware (Tenstorrent), serving open-weight LLMs and multimodal models to customers through our own API platform. We’re a small team building everything from the request gateway to the tooling that keeps the hardware healthy, so interns work on real production systems from week one.

The role

This is a flexible role. Depending on what the cluster needs and where your interests lie, you’ll rotate across:

API gateway & request path

• Build and improve our inference gateway in Go: routing, auth, API keys, rate limiting and quotas
• Handle streaming responses (SSE/WebSockets) for text and audio, and file/URL-based responses for images and video
• Input validation, request queuing, load balancing across nodes and models

Observability & metrics

• Build services that collect node, accelerator and model-level metrics (throughput, latency, tokens/sec, utilization, errors)
• Instrument services with Prometheus/OpenTelemetry and build Grafana dashboards and alerts
• Usage metering and per-customer token accounting for billing

Internal tools & cluster operations

• CLI tools and internal services for deploying models, health checks and node management
• Automation for provisioning, upgrades and incident response
• Whatever else it takes to keep the cluster running reliably

What we’re looking for

• Solid fundamentals in Go (goroutines, channels, context, error handling, testing), or strong skills in another language plus a willingness to get productive in Go quickly
• Understanding of HTTP, REST APIs and basic networking
• Comfortable on Linux and the command line
• Familiarity with Git and writing clean, tested code
• Curiosity about how AI models are actually served in production

Nice to have

• Experience with Docker / Kubernetes
• Prometheus, Grafana or any observability stack
• gRPC, Redis, PostgreSQL, message queues
• Exposure to LLM serving (vLLM, OpenAI-compatible APIs, tokenizers)
• Open-source contributions or personal projects in Go

What you’ll get

• Hands-on work with a production AI cluster, which very few internships offer
• Direct mentorship from the engineers who built and run it
• Ownership of real features that ship to customers
• [PPO opportunity / certificate / flexible hours / other perks]