Software Engineer, AI
Posted 6hrs ago
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
AI-focused Python Software Engineer building FastAPI services and generative AI integrations. Securing and operating CodiLime’s multi-tenant enterprise SaaS platform.
Responsibilities:
- Design, build, test and operate production Python services with FastAPI, including endpoint design, Pydantic validation, authentication, error handling and observability
- Write and optimize SQL against PostgreSQL and contribute to schema and data-model decisions
- Deploy and run services on Kubernetes, including reasoning about stateful versus stateless workloads and persistent data
- Build and integrate generative AI features connecting LLMs with enterprise data, APIs and internal tools
- Design prompts, tool calls, workflow orchestration, context management, guardrails and human review
- Evaluate AI workflows and monitor quality, latency, cost and failure rates
- Maintain automated pytest coverage and help build automated quality gates
- Use AI coding assistants while validating generated output before production
- Apply security-first practices including tenant isolation, RBAC and least-privilege data access
- Participate in code reviews and improve engineering practices
- Document data flows, API contracts and AI workflow behavior for non-engineers
- Collaborate with product managers, UX designers, engineers, data scientists and client-facing stakeholders
Requirements:
- 6+ years of professional experience in software engineering with Python
- Experience with FastAPI and Pydantic - endpoint design, dependency injection, models and validators, auth and error handling
- Solid REST API design - versioning, predictable error semantics, token management, retries - with OAuth 2.0/Okta, JWT and role-based access control (RBAC)
- Experience with automated testing with pytest
- Strong SQL fundamentals (joins, GROUP BY, aggregate functions, query optimisation) and hands-on experience using PostgreSQL from Python - ORMs, database drivers, and a clear understanding of the trade-offs between synchronous and asynchronous drivers
- Solid experience deploying and operating applications on Kubernetes
- Experience with CI/CD pipelines (ideally GitHub Actions)
- Hands-on experience integrating LLM APIs into production applications, including prompt design, tool/function calling, and safe handling of non-deterministic output
- Experience using AI coding assistants such as Claude Code, Codex, or similar on a daily basis
- Understanding of multi-tenant SaaS application security
- Ability to evaluate AI features and monitor quality, latency, cost, and reliability
- Product mindset and ownership - you identify problems, propose solutions, and take features from idea to production
- Strong communication skills and good knowledge of English (minimum C1 level)
- Evaluation and observability tooling for LLM features: eval harnesses, tracing, output scoring (LangSmith, Langfuse, Arize Phoenix or OpenTelemetry) — nice-to-have
- Systematic prompt optimisation — nice-to-have
- Agentic patterns beyond single LLM calls: multi-step orchestration, context management, guardrails and human-in-the-loop — nice-to-have
- Retrieval-augmented generation (RAG) and vector search (pgvector, Pinecone), plus frameworks such as LangChain or LangGraph — nice-to-have
- Experience with Snowflake or a comparable cloud data warehouse — nice-to-have
- Temporal, Socket.IO / WebSockets, Redis, pub/sub — nice-to-have
- Working knowledge of React / TypeScript — nice-to-have
Benefits:
- Flexible working hours and approach to work: fully remotely, in the office or hybrid
- Professional growth supported by internal training sessions and a training budget
- Solid onboarding with a hands-on approach to give you an easy start
- A great atmosphere among professionals who are passionate about their work
- The ability to change the project you work on



















