About the Role
Responsibilities
- Build proxy infrastructure from scratch. Not integrate existing solutions. Write the core systems yourself.
- Design and implement rotation systems. IP rotation, session management, sticky sessions, intelligent routing.
- Manage IP reputation and quality. Monitor, track, score, and optimize IP health across pools.
- Build bandwidth reporting and metering. Track usage, report stats.
- Write reverse proxy systems. High-performance, optimized for our use case. Not just configure squid or nginx.
- Handle multiple proxy types. Residential, mobile, datacenter, ISP. Understand the differences, characteristics, and when to use each.
- Implement geo-targeting. Country, city, ASN level targeting.
- Optimize for performance. Every millisecond matters. Every connection matters.
- Work with networking fundamentals. Routing, DNS, TCP/UDP tuning, connection pooling, concurrent connection limits.
- Support multiple protocols. SOCKS5, HTTP, HTTPS. Different session types and authentication methods.
- Understand HTTP/2 and HTTP/3 differences. And how they affect proxy behavior.
- Build monitoring and health checks. Track proxy pool health, failover, redundancy.
- Handle IPv4 and IPv6. Understand the differences and implications.
Requirements
- Deep networking knowledge. TCP/UDP, sockets, connection handling, routing. Not surface-level.
- Protocol expertise. HTTP, HTTPS, SOCKS5. You understand them at the byte level.
- Can write a reverse proxy from scratch. Not just configure squid or nginx. Write the proxy.
- Performance obsessed. You profile, measure, and optimize. You know where the bottlenecks are.
- Understand proxy types. Residential, mobile, datacenter, ISP. Their characteristics, use cases, and limitations.
- Experience with IP reputation. How it works, how to manage it, how to recover from blacklisting.
- Geo-targeting knowledge. Country, city, ASN level. How it works, accuracy limitations.
- Linux networking. iptables, routing tables, network namespaces. Comfortable at this level.
- Self-directed. You don't need someone to tell you what to build next.
- Great writer. Documentation, architecture decisions, protocol specs. Clear communication matters.
- Fluent in English, written and spoken.
- BSc or higher in CS or equivalent technical field.
- Built proxy or networking infrastructure before (not just used it).
- Deep understanding of HTTP/HTTPS/SOCKS5 protocols.
- Experience with high-throughput, low-latency systems.
- TCP/UDP programming at socket level.
Nice to Have
- Experience building residential or ISP proxy networks.
- Understanding of ISP relationships and mobile carrier specifics.
- IP rotation and session management systems.
- IP quality scoring and reputation management.
- Geo-targeting implementation (country, city, ASN).
- Multi-tenancy and customer isolation.
- HTTP/2 and HTTP/3 proxy experience.
- Terraform for infrastructure provisioning.
- Experience at massive scale.
Benefits
- Fully Remote. Work from anywhere.
- Equity share. Own a piece of what you're building.
- Profit sharing. When we win, you win.
- Greenfield project. Build something from scratch.
- Annual team retreats (last one in Spain).
Work Arrangement
Remote (Worldwide)
Team
Structure: lean, profitable, bootstrapped team
Additional Information
- We use existing solutions today. We want someone who understands their limitations and can build something better.
- You define the architecture. We have requirements, not a blueprint. You'll decide how to build it.
- No existing codebase. You're building from zero. No one to ask how it was done before.
- Performance is everything. Slow proxies are useless proxies. You'll obsess over latency and throughput.
- Complexity at scale. Managing thousands of IPs, tracking reputation, handling failures gracefully.
- Deep debugging. When something breaks at the network level, you need to figure it out. Packet captures, connection traces, protocol analysis.
- Understanding trade-offs. We use existing solutions now. You should understand what they do well, where they fall short, and what could be done better.
- If you prefer working with existing tools and frameworks without understanding what's underneath, this isn't for you.
- We ship faster because we use AI tools aggressively. Cursor and Claude. But you own the output. If you can't evaluate what AI produces, you're not ready for this role.
- Include "Octopus Tentacle" in your application so we know you read this.