Senior AI Engineer, Agentic Systems (Client-Facing) — B2B Contract, Remote

PowerMind Contractors; Polska; Hybrid; PLN 239,000 - 292,000; Full time

Benefits offered by this job

Remote work

Job summary

PowerMind Contractors is seeking a Senior AI Engineer with deep agentic and RAG experience to join a boutique IT consultancy delivering AI and modernization projects for enterprise customers across the UK and Europe.

The role is fully remote on a B2B, project-based engagement with rates of $45-55 per hour depending on experience. You will own how the AI behaves, build agent workflows, tune retrieval, ensure production readiness, and communicate with the client in plain language.

Qualifications

• 6+ years in software engineering and expert-level Python.
• 2+ years building LLM-based systems that reached production with real users.
• Production experience with Haystack or LangChain/LangGraph.
• Hands-on RAG and agentic systems, including loop control and multi-source retrieval.
• Experience with at least one vector database.
• Evaluation and tracing practice: you can show how you proved a change improved the system.
• Docker and CI/CD fundamentals.
• Direct client-facing experience: workshops, requirement discussions, presenting trade-offs.

Responsibilities

• Design and implement agent workflows with planning, tool use, multi-step reasoning and routing.
• Define stopping criteria for reasoning loops and manage state across runs.
• Design tool calls and enforce structured outputs; package services with Docker and deploy to Azure.

Skills

Python

LLM-based systems

Agentic systems

RAG concepts

Multi-source retrieval

Client-facing

Production-grade code

Docker

Haystack

LangChain/LangGraph

Vector databases

Azure

CI/CD

Tools

Haystack

LangChain/LangGraph

Vector databases

Azure

CI/CD pipelines

Job description

We're looking for a Senior AI Engineer with deep agentic and RAG experience to join our client, a boutique IT consultancy delivering AI and modernization work for enterprise customers across the UK and Europe.

Fully remote, B2B, project-based engagement months with a rate of $45-55 per hour depending on experience. Strong delivery on the first project opens the door to follow-on work.

The role

You'll join a small delivery team as the engineer who owns how the AI actually behaves. An architect is already on the project, so this is not an architecture seat. You build the agent workflows, tune the retrieval, prove the quality and get it running in production, and you do it while sitting in front of the client.

This is an engineering role with client exposure, not a strategy or advisory role. You write code daily and you also explain, in plain language, why the system answered the way it did. The core technical challenge is reasoning over many heterogeneous data sources: how to combine them, how the reasoning loop should behave, and when to stop and return a concrete answer.

What you'll do

Agentic systems and orchestration

• Design and implement agent workflows in Haystack or LangChain/LangGraph: planning, tool use, multi-step reasoning, routing between sources
• Define stopping and exit criteria for reasoning loops: confidence thresholds, iteration and token budgets, fallback paths, human-in-the-loop points
• Design tool and function calling, enforce structured outputs, manage state across multi-step runs

Retrieval and data

• Build retrieval across multiple sources: chunking, embeddings, hybrid search, reranking, graph-based retrieval
• Merge, deduplicate and reconcile results when sources disagree, with traceable citations
• Work with vector databases and graph databases as first-class parts of the pipeline
• Apply vision-language models where the input is visual: scanned documents, diagrams, images

Quality and production

• Build evaluation for retrieval and answer quality, with regression checks when prompts or models change
• Instrument the system: tracing of agent steps, latency, token usage and cost per request
• Package services with Docker and deploy to Azure, working with existing CI/CD pipelines
• Handle the boring production parts properly: retries, timeouts, guardrails, error handling
• Run technical sessions with the client and answer hard questions in real time
• Translate business requirements into system behaviour and back again
• Flag risks early: hallucination, data access limits, cost blow-ups, latency

Must have

• 6+ years in software engineering, expert-level Python
• 2+ years building LLM-based systems that reached production with real users
• Production experience with Haystack or LangChain/LangGraph
• Hands-on RAG and agentic systems, including loop control and multi-source retrieval
• Experience with at least one vector database
• Evaluation and tracing practice: you can show how you proved a change made the system better
• Docker and CI/CD fundamentals
• Direct client-facing experience: workshops, requirement discussions, presenting trade-offs

Nice to have

• Graph databases (Neo4j or similar), GraphRAG or knowledge-graph-backed retrieval
• VLM and multimodal RAG experience (significant plus)
• MCP or A2A
• Prior consultancy, agency or contracting background

Details

• B2B contract via PowerMind Contractors
• 100% remote, CET and UK working hours
• 2-3 month engagement, extension and follow-on projects possible
• You'll be working as part of our client's delivery team, under their brand, with direct contact with the end customer
• Background check and hardware security key (YubiKey) required before start