# Arc.dev (client undisclosed) — Senior AI Product Engineer

| Field | Value |
|---|---|
| **Date found** | 2026-06-02 |
| **Company** | Arc.dev (client undisclosed) |
| **Role** | Senior AI Product Engineer |
| **Location** | EMEA Remote |
| **Salary** | Undisclosed |
| **Job URL** | https://www.linkedin.com/jobs/view/4421807338/ |
| **Status** | New |

---

## Company Research

Recruiting agency — client undisclosed.

---

## Job Summary

**What they do:** An unnamed company building an AI-powered "answer engine" — a production RAG system that reasons, verifies facts, and handles complex queries at low hallucination rates.

**The role:** Senior AI Product Engineer owning the full RAG pipeline and agentic reasoning layer for the answer engine's production application.

**Core work:**
- Build RAG pipelines using LangChain/LangGraph with hybrid search (pgvector, Qdrant) and re-ranking models
- Implement agentic tools (web search, code execution) with robust error handling and state management
- Develop evaluation loops to verify citation accuracy and answer relevance

**Stack:** Python · LangChain · LangGraph · pgvector · Qdrant · Pinecone · RAG · agentic AI · multi-step reasoning

**Work style:** Fully remote EMEA; placement via Arc.dev talent marketplace

---

## Score: 77%

| Dimension | Score | Justification |
|---|---|---|
| Agentic AI depth (25%) | 75% | RAG + agentic tool use, multi-step reasoning, LangChain/LangGraph explicit — solid agentic depth for an answer engine |
| Tech fit (25%) | 85% | Python, LangChain, LangGraph, vector DBs — excellent stack overlap with Luca's toolset |
| Remote fit (25%) | 90% | Fully remote EMEA — ideal |
| Company culture fit (15%) | 50% | Client unknown — can't assess culture; Arc.dev suggests tech-forward startup |
| IC/leadership balance (10%) | 90% | Purely IC — individual contributor ownership of a production AI system |
| **Final (weighted)** | **77%** | |

---

## Strengths

- Strong LangChain/LangGraph stack match — direct overlap with Luca's agentic AI expertise
- Fully remote EMEA — perfect location fit
- Agentic tool use, multi-step reasoning, RAG evaluation frameworks — genuinely production-grade work
- Only 48 applicants on day 1 — low competition for a high-quality role

---

## Weaknesses & Risks

- Client is anonymous via Arc.dev marketplace — can't assess culture, team size, or company stability
- Salary undisclosed — significant risk given €100k floor requirement
- 9+ years experience required — strict bar
- "Answer engine" is a narrow problem domain, not a full agentic AI platform

---

## Suggestions

- Emphasise production RAG + agentic tool calling experience (cybersecurity platform, Fenergo pipelines)
- Highlight LangGraph multi-agent orchestration work and evaluation framework design
- Ask recruiter upfront: who is the client? What stage is the company? What's the comp range?
- Stress ownership of full system lifecycle including evaluation loops and retrieval quality measurement

---

## Interview Tracker

| Stage | Date | Notes |
|---|---|---|
| Expired | 2026-06-04 | Posting removed — no longer accepting applications |
| Applied | | |
| Recruiter screen | | |
| Technical interview | | |
| Final round | | |
| Offer / Outcome | | |