#### About the company
Relei is an AI-native platform for business development and project intelligence — helping consultancies turn their project archives into sales-ready assets like case studies, proposals, and BD decks.
We’re currently in active pilots with digital and design agencies, and our previous engineer is handing over the product this week.

We’re now looking for an AI Product Engineer who can immediately take over and extend our agentic backend — someone who’s comfortable owning both hands-on engineering and intelligent orchestration design.

#### You’ll Work On
- Extend Relei’s multi-agent orchestration system built with FastAPI, LangGraph, LangFuse, and AWS App Runner.
- Build and refine retrieval, context assembly, and structured generation workflows for use cases like proposal and report generation.
- Implement integrations with Google Drive, Slides, Slack, and other team tools.
- Optimize prompts and context construction for different LLMs (Azure OpenAI, GPT-4, Together, etc.).
- Improve evaluation and observability using LangFuse, tests, and metrics.
- Collaborate directly with the founder on rapid iteration and pilot delivery.

#### Core Tech Stack
- Backend: Python (FastAPI, Poetry, Pytest)
- AI Orchestration: LangChain / LangGraph / LangFuse
- Infra: Docker, AWS App Runner, ECR, CloudWatch
- Integrations: Google Drive & Slides API, Slack API
- LLMs: Azure OpenAI, GPT-4, TogetherAI (multi-provider setup)

#### Ideal Profile
- Strong Python engineer with 3–6 years of experience building LLM-based products.
- Hands-on experience designing or maintaining agentic AI workflows — ideally using LangChain / LangGraph.
- Familiar with context engineering, retrieval pipelines, and structured generation.
- Fluent in debugging and deploying containerized apps (Docker → AWS).
- Strong bias for shipping: pragmatic, resourceful, and thrives in lean setups.
- Excellent async communicator; able to work autonomously.

#### Bonus:
- Experience with LangFuse or prompt evaluation frameworks.
- Familiarity with browser extensions (React/TS) or Slack bots.
- Interest in joining as a Founding Engineer post-contract.

#### Engagement
- Contract: Freelance (full-time, immediate start)
- Duration: ~6 weeks to complete 2 pilot deliveries (mid-Oct to Nov)
- Availability: Immediate start required
- Location: Remote, CET-friendly

#### You’ll Work On
1. Chemistry Pilot – Proposal Generation
Finalize PoV + Scoping Agents
Complete Deck Agent that fills Google Slides templates (text, tables, team CVs).
Deliverable: Draft proposal deck in client’s brand template.
Due: 16 Oct 2025

2. Cord Pilot – Campaign Report Automation
Build new workflow that turns shared Drive data (images, metrics, feedback) into branded HTML emails.
Deliverable: Polished HTML campaign report ready for Slack / Gmail delivery.
Due: 24 Oct 2025

3. Stabilization & Handover
Fix bugs, refine outputs, document setup, record short Loom walkthroughs.
Due: 15 Nov 2025

#### Collaboration Model
- Works directly with the founder on Slack + GitHub.
- Optional async access to previous tech lead for architecture advice.
- Supported by a QA / AI content intern for reviewing outputs and testing workflows.
- Customers interact only with founder; freelancer stays focused on tech delivery.