Lead AI Engineer, Agent Platform & LLM Systems

Nacre Capital; Turkey; Remote; TRY 1,000,000 - 1,800,000; Full time

Benefits offered by this job

Direct line to CTO

Budget for models and tooling

Remote-first, globally distributed

Competitive compensation

Job summary

EveryWatch is seeking aSenior AI Engineer, Agents - an Agent Architect who will design, build, and scale AI-powered solutions across EveryWatch. You will own the agent roadmap, implement sub-agents, and ensure a cost-effective, scalable platform integrated with watch-market data.

You will work closely with engineering, product, sales, and data teams, shipping production-grade agents, optimizing latency and model costs while advancing the ecosystem of intelligent agents for collectors, dealers,

Qualifications

• 6+ years shipping production software, with 2+ years on LLM systems.
• Real agentic depth: ReAct or equivalent loops, tool/function calling, planners, state & checkpointing, long-term memory, HITL steps, streaming.
• Hands-on LangGraph plus a tracing/observability stack (LangSmith, Langfuse or similar).
• Python in production: FastAPI, async job processing, clean service boundaries.
• RAG done properly: retrieval + re-ranking + relevance judgement, and honest evaluation of all three.
• Evaluation as a habit: RAGAS/DeepEval/GEVAL, LLM-as-judge, pass@k.
• Cloud production experience - AWS (Bedrock, SQS, EC2/EKS, S3) or equivalent - with Docker and CI/CD.
• We explicitly want someone who asks why you built it like that?

Responsibilities

• Invent the agent roadmap with sales, data and product; rank opportunities by feasibility, cost and impact.
• Build and deploy agents yourself: architecture, code, tools, memory, sub-agents, evaluation harness.
• Own the agent platform: shared tool layer on EverWatch data, MCP surface over backend, memory, tracing, eval harness.
• Harden WatchChat: multi-turn state and memory, latency, cost, tool-call reliability, regression gates.
• Make quality measurable: golden datasets, verifiers for tool correctness, LLM-as-judge on held-out sets.
• Treat cost and latency as design constraints; use cheap models for routing and strong models where needed; caching and fine-tuning as appropriate.

Skills

LLM systems

Agent platform design

Python production

Memory & state management

Evaluation & verifiers

Cost and latency as design constraints

Distributed systems

Observability / tracing

Tools

LangSmith

Langfuse

LangGraph

Docker

CI/CD

Job description

EveryWatch is seeking aSenior AI Engineer, Agents - an Agent Architect who will design, build, and scale AI-powered solutions across EveryWatch. You will own the agent roadmap, implement sub-agents, and ensure a cost-effective, scalable platform integrated with watch-market data.

You will work closely with engineering, product, sales, and data teams, shipping production-grade agents, optimizing latency and model costs while advancing the ecosystem of intelligent agents for collectors, dealers,