Senior Python AI Engineer (Agentic Systems / LLM)
New
B
BritenetAI systems
Workplace type: remote; Country code: PL, Daily overlap with US time zones (CET afternoon), usually 9-5Full-TimeSenior
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Languages
- Fluent written and spoken English
- Experience
- 5+ years of experience in software development, mostly in Python; 2+ years building LLM applications used in production
- Required Skills
- AWSPythonFastAPIRESTful APIs
Requirements
- Have 5+ years of software development experience, mostly in Python.
- Have 2+ years building LLM applications used in production.
- Build backend services and REST APIs using FastAPI, Flask, or Django.
- Use pytest for automated testing, and apply type hints, static analysis, and linting with mypy and ruff, as well as CI.
- Work across the stack, including frontend changes.
- Use coding agents such as Claude Code, Cursor, or Codex daily; write task specifications and review agent output with tests and data.
- Configure and maintain agent harness components, including project instructions, skills, hooks, subagents, and MCP servers.
- Build tool-calling and multi-step agent workflows, combining rule-based logic with LLM components and using structured outputs such as Pydantic.
- Work with RAG, embeddings, vector and hybrid search, and reranking.
- Design evaluations on labelled data, track regressions, and use LLM observability, including cost and latency monitoring.
- Have experience with AWS, Azure, or GCP, including managed AI services.
- Work independently, take ownership, and work directly with client domain experts to translate requirements into technical solutions.
- Communicate technical decisions to non-technical stakeholders and have fluent written and spoken English.
- Work remotely with daily overlap with US time zones (CET afternoon), usually 9-5.
Responsibilities
- Build production GenAI systems that support domain experts.
- Develop agentic decision-support systems for document review, including document extraction and traceability to source pages.
- Implement review rules using deterministic logic and LLM components, and address model nondeterminism.
- Build evaluation harnesses that compare system output with expert decisions.
- Develop retrieval and answer APIs with hybrid search, reranking, citations, guardrails, and ingestion pipelines.
- Contribute to LLM-enabled research platform workflows, including source libraries, deduplication, drafting, review, and approval.
- Build custom tools on an internal LLM platform, data pipelines, and market intelligence analytics.
- Configure and maintain the coding-agent harness, including project instructions, skills, hooks, subagents, and MCP servers.
- Design evaluations on labelled data, track regressions, and monitor LLM cost and latency.
View Full Description & ApplyYou'll be redirected to the employer's site