Crowdcore Inc. / QuotaFlow · [Location / Remote] · Up to $200K base + a strong performance bonus

About usWe’re an AI-native company with a small team and a large business. We build and run QuotaFlow, an LLM gateway that routes Claude, GPT and Gemini traffic for engineering teams and enterprises. It handles routing, failover, prompt caching, quotas, billing and usage dashboards, all on AWS. A handful of engineers run production traffic for some of the heaviest AI users anywhere, so each person here has a lot of leverage.

AI agents aren’t a side tool for us; they’re how we build. Our repo has an entry-point file that agents and humans both follow. Work runs in parallel agent lanes on isolated worktrees. A separate reviewer agent tries to disprove each change before it merges, and releases go out on scheduled trains with canary deploys. We want someone who will make this system better, not just use it.

What you’ll do

- Own the hot path: request routing across many upstream providers, failover, cache-aware stickiness, streaming and rate limits.
- Turn business requests into working systems fast: onboarding a new provider, custom customer pricing, invoicing, usage reports. Often the same day.
- Build the harness: instructions, scripts, probes, test selectors and review loops that let Codex and Claude Code do reliable work on a large production codebase.
- Debug production from data: trace latency spikes, cache-hit drops and billing gaps to a root cause, then explain it in plain language.
- Ship the frontend too:



dashboards and admin tools, with AI doing most of the typing.

What we’re looking for
- You use Codex, Claude Code or similar tools every day, and you’ve gone past chatting with them. You’ve set up the instructions, worktrees, tests, review loops and automation that make agents productive and keep their output correct.
- Solid backend fundamentals: APIs, Postgres, concurrency and distributed-systems trade-offs. Go is a plus.

- Hands-on AWS experience (ECS/Fargate, IAM, CloudWatch and the like).
- You’re comfortable shipping frontend with AI help. You don’t need to be a frontend specialist.
- No ego. You listen, ask the right question, work out what the business actually needs, and say “I was wrong” quickly.
- Clear written communication.
- A player. You want to own outcomes, not close tickets.

Nice to have
- Infrastructure experience at high traffic: high-throughput proxies or gateways, long-lived streaming (SSE), load balancing, backpressure and cost-aware scaling.
- Knowledge of how LLM APIs work under the hood: token accounting, prompt caching, provider rate limits and error semantics.

Why join
- Pay: up to $200K base plus a great bonus tied to how the business does.
- Leverage: small team, big revenue. Your work moves real numbers right away.
- AI-native for real: we use agents to build the product, and the product is AI infrastructure.
- Speed: no layers.

You talk directly to the people making decisions.

📌 AI Software Engineer (AI-native · Backend + Infra) (Toronto)
🏢 quotaflow
📍 Toronto