Back-end AI Engineer Python / LLM Orchestration

Worldwide | Aug. 31, 2026

Report as Closed

Company: Improving

Country: Worldwide

Type: Remote

Employment: Full-time

Description: Improving is an IT services firm focused on AI, Data, and Applications. We modernize legacy systems, build cloud-native platforms, and deliver future-ready solutions through collaborative, long-term partnerships that drive measurable outcomes.

✓ Python at production level. Clear possession of orchestration and business logic, not scripting

✓ Experience with multi-agent frameworks / LLM: Hands in Pydantic AI, LangChain, CrewAI, or similar (structured definitions of agents, workflow orchestration, tool-calling)

✓ RAG implemented in the practice: Retrieval pipelines, embeddings, vector search, chunking strategies (bonus if code-aware)

✓ Azure-native stack fluency: Azure SQL, Cosmos DB, compute on Azure — real familiarity, not theoretical

✓ API integration skills: The backend communicates cleanly with the .NET/React frontend; able to work autonomously on that border

✓ Solid spoken and written English: Probable client-facing exposure; stakeholders in the USA

✓ Time overlap with US Eastern: Standups and client touchpoints require alignment

• Design and maintain multi-agent orchestration: Implement complex decision flows between LLM agents using frameworks such as Pydantic AI, LangChain or CrewAI
• Manage RAG pipelines: Develop efficient retrieval strategies, configure vector search and optimize chunking for code queries
• Integrate with Azure: Work natively with Azure SQL, Cosmos DB and compute services — without abstractions that disguise the platform
• Clean APIs: Expose functionality clearly and without friction to the frontend
• Observability and debugging: Instrument agent decisions, track bugs, iterate quickly based on real telemetry

• Long-term contract.
• 100% Remote.
• Vacation and PTOs
• Possibility of receiving 2 bonuses per year.
• 2 salary reviews per year.
• English classes.
• Apple equipment.
• Online course platform
• Budget for purchase of books.
• Budget for purchasing work materials
• much more..

• AI Observability: Pydantic Logfire or similar (decision traceability, success/failure monitoring)
• AI-assisted dev tools: Cursor or similar (client uses this; expected speed reflects that)
• Background in consulting / delivery: Ambiguous and fail-fast environments; better ramp than fully-spec'd tickets
• Multi-cloud exposure: AWS (Bedrock, Lambda, document analysis) — not necessary here, but adds for flexibility in other projects
• Industrial domain: Manufacturing, construction, or ERP-related experience — BlueScope has real logic (custom steel components, pre-detailing flows); accelerates ramp
• Static analysis / codebase parsing: Direct to the core of Chiron (extraction of workflows from source code)

Apply here:

Web: Apply here

Emails: