Senior Software Engineer — AI Infrastructure / Backend Systems (Go, Python, React

Overview We are looking for a strong Software Engineer / Backend Engineer to help us with a small, well-scoped engineering task within Cortex, an AI infrastructure platform currently under active development. Cortex is designed as a unified AI gateway and orchestration layer that provides a consistent interface for interacting with multiple AI providers and models. The platform involves provider abstraction, model routing, streaming responses, embeddings, cost tracking, budgets, reliability controls, and production-grade infrastructure. This is initially a small paid task intended to evaluate engineering quality, codebase familiarity, problem-solving ability, and attention to production concerns. There is potential for additional work if the collaboration is successful. We are specifically looking for someone who is comfortable working in an existing production-oriented codebase rather than someone who only builds greenfield applications. --- What You'll Work On Your initial assignment will involve making a targeted change within the Cortex codebase. The task will require you to: - Understand an existing backend architecture - Navigate an unfamiliar codebase efficiently - Identify the correct abstraction/layer for a change - Implement the requested functionality cleanly - Maintain existing interfaces and contracts - Write or update appropriate tests - Consider error handling and edge cases - Ensure the implementation does not introduce regressions - Provide a concise explanation of your implementation and design decisions The initial task is deliberately small, but the codebase contains more complex infrastructure that you may work on later. --- Required Technical Skills Backend Engineering Strong experience with backend software engineering is required. You should be comfortable with: - REST APIs - HTTP services - API design - Request/response lifecycle - Middleware - Authentication and authorization concepts - Error handling - Context propagation - Concurrency - Resource management - Graceful shutdown - Configuration management - Structured logging - Service-to-service communication - Database-backed applications --- Go Strong Go/Golang experience is highly preferred. You should be comfortable working with: - Go modules - Interfaces - Structs - Goroutines - Channels - Contexts - HTTP servers and clients - "net/http" - Error wrapping and propagation - Dependency injection - Concurrency control - JSON serialization - Unit testing - Table-driven tests - Clean package architecture You should understand idiomatic Go rather than simply being able to write syntactically correct Go. Experience designing interfaces and abstractions is particularly important. --- Python Experience with Python backend development is also required or strongly preferred. Relevant experience includes: - FastAPI - Pydantic - Async Python - HTTP APIs - Background processing - API integrations - Python testing - Service-oriented architectures Experience with the Model Context Protocol (MCP) or AI tool/runtime infrastructure is a strong advantage. --- AI / LLM Infrastructure Experience working with modern AI/LLM systems is highly desirable. You should understand concepts such as: - LLM APIs - Chat/completions APIs - Streaming responses - Embeddings - Model providers - Provider abstraction - Model routing - AI gateways - Token usage - AI cost tracking - Rate limits - Retries - Timeouts - Fallbacks - Provider failures - Model capabilities - Multi-provider architectures Experience integrating providers such as: - OpenAI - Google Gemini - Anthropic - OpenRouter - Groq - Mistral - Cerebras - Other OpenAI-compatible APIs is a strong advantage. --- Software Architecture We are looking for someone who can reason about architecture, not just implement tickets. You should understand: - Clean architecture principles - Domain-driven design concepts - Separation of concerns - Dependency inversion - Interface-based design - Service boundaries - Repository patterns - Provider abstractions - Domain/application/infrastructure separation - API contracts - Backward compatibility - Extensibility You should be able to look at an existing system and determine where a change belongs instead of simply modifying the first file that appears relevant. --- Databases Experience with relational databases is expected. You should be comfortable with: - PostgreSQL - SQL - Database schema design - Indexes - Transactions - Constraints - Query optimization - Connection management - Migrations - Atomic database operations Experience with: - PostgreSQL-compatible cloud databases - Redis - Caching - Distributed locking - Usage/cost tracking is beneficial. --- Frontend Frontend experience is a plus because Cortex has supporting administrative and operational interfaces. Experience with: - React - TypeScript - JavaScript - REST API integration - Component architecture - State management - Modern frontend tooling is desirable. You do not need to be primarily a frontend developer. Backend and systems engineering experience is the priority. --- Testing & Quality Production-quality engineering is important to us. You should be comfortable writing: - Unit tests - Integration tests - API tests - Regression tests - Edge-case tests You should understand: - Test isolation - Mocking - Test fixtures - Error-path testing - Concurrency testing where appropriate - Regression prevention We value tests that verify behavior and contracts, rather than tests written purely to increase coverage numbers. --- Production Engineering Experience operating real applications in production is strongly preferred. You should understand: - Docker - Linux - Environment configuration - Process management - CI/CD - Reverse proxies - Observability - Structured logging - Metrics - Health checks - Timeouts - Retries - Resource limits - Failure recovery - Deployment safety Experience with cloud platforms such as AWS, DigitalOcean, Cloudflare, or similar platforms is a plus. --- Code Quality Expectations We care significantly about engineering quality. Your code should be: - Idiomatic - Readable - Maintainable - Testable - Explicit - Well-structured - Minimal where possible - Consistent with the existing architecture Avoid: - Unnecessary abstractions - Large unrelated refactors - Copy-pasted logic - Hidden global state - Poor error handling - Silent failures - Breaking existing contracts - Over-engineering a small requirement When modifying an existing system, we expect you to work with the architecture rather than fight it. --- Git & Development Workflow You should be comfortable with Git and collaborative development. Expected experience includes: - Git - Feature branches - Pull requests - Meaningful commits - Reviewing diffs - Rebasing/merging - Resolving conflicts - Writing useful commit messages - Keeping changes focused For the initial task, we expect a clean and reviewable change set. --- What We Value Beyond technical skills, we are particularly interested in engineers who: - Think before coding - Can quickly understand unfamiliar code - Ask precise questions when requirements are ambiguous - Can independently investigate problems - Care about failure modes - Understand system boundaries - Write maintainable code - Take ownership of their changes - Communicate clearly - Can explain technical decisions - Prefer simple solutions when simple solutions are sufficient You should be comfortable receiving a task such as: «"This component currently behaves incorrectly under this condition. Investigate the existing implementation, identify the appropriate layer to fix it, implement the correction, add regression coverage, and explain your approach."» rather than requiring extremely detailed instructions for every implementation step. --- Initial Task The initial engagement will be a small paid engineering task in the Cortex codebase. The task will be intentionally scoped so that we can evaluate: 1. How quickly you understand the existing architecture 2. Your ability to navigate an unfamiliar codebase 3. Your ability to identify the correct implementation point 4. Code quality 5. Testing discipline 6. Error handling 7. Attention to existing contracts 8. Communication 9. Ability to avoid unnecessary changes If the initial task goes well, there is potential for continued development work on Cortex and related AI infrastructure. --- Ideal Candidate The ideal candidate is someone with approximately 3+ years of professional software engineering experience, with strong backend experience and exposure to AI/LLM systems. A particularly strong candidate will have experience with: Go + Python + PostgreSQL + Redis + React/TypeScript + Docker + REST APIs + LLM APIs and understands how these technologies fit together in a production system. Experience building: - AI gateways - LLM platforms - Developer infrastructure - API platforms - SaaS backends - Distributed systems - Data/knowledge pipelines - MCP servers/runtimes - AI orchestration systems would be a major advantage. --- Nice-to-Have Experience Any of the following will strengthen your application: - MCP / Model Context Protocol - AI agent infrastructure - RAG systems - Vector databases - Embedding pipelines - Knowledge graphs - LLM observability - AI cost optimization - Multi-provider AI systems - OpenAI-compatible APIs - Streaming architectures - Event-driven systems - Distributed systems - Cloudflare Workers - Cloudflare R2 - AWS - DigitalOcean - Neon/PostgreSQL - Redis - Prometheus/OpenTelemetry - CI/CD - Security engineering --- Engagement Project type: Small paid technical task, with potential for ongoing work Initial scope: One focused engineering task in an existing AI infrastructure codebase Potential follow-up: Additional Cortex development, architecture work, bug fixes, hardening, testing, and new features Communication: Async-friendly, with clear written technical communication expected --- How to Apply Please include the following in your proposal: 1. Brief introduction Tell us about your backend/software engineering experience. 2. Relevant experience Describe one project where you worked on a system involving: - Go or Python - APIs - AI/LLM infrastructure - PostgreSQL - distributed systems - or another technically complex backend system. 3. Architecture experience Briefly describe a situation where you had to modify an existing codebase without breaking its architecture or existing contracts. 4. Code sample Provide a link to a relevant GitHub repository, portfolio project, or other code sample that demonstrates your engineering ability. If the code is private, briefly describe the project and your specific contribution. 5. Availability Tell us: - Your current availability - Approximate hours per week - Preferred working arrangement - Your hourly rate 6. Short technical question In your own words: When you are given a small feature request in an unfamiliar backend codebase, what is your process for determining where the change should be implemented? We are more interested in your reasoning than a textbook answer. --- Important This is not a code-generation-only position. We are looking for an engineer who can understand an existing system, reason about its architecture, make a targeted change, test it properly, and communicate the result clearly. The initial task is small, but the quality bar is high. If you demonstrate strong engineering judgment on the initial task, there is an opportunity for ongoing work on Cortex and other AI infrastructure projects.