AI Engineer

WHO WE ARE

DELVE Deeper is a global performance media agency where data, technology, and marketing intersect.

We help brands like UNICEF, Virgin Voyages, and Orange grow by using data, analytics, and automation to drive measurable results. Our teams work at the intersection of media, data science, and technology in a fast-paced, international environment.

ROLE OVERVIEW

Dedicated individual contributor focused entirely on building and maintaining a multi-agent AI system that automates performance media trading decisions.

This is not a generalist developer role. You will work within a complex multi-agent codebase, operate under strict evaluation-first protocols, and develop deep understanding of the business domain to make sound implementation decisions without constant oversight.

Data errors have direct financial consequences for clients — engineering quality is non-negotiable.

CORE RESPONSIBILITIES

• Agent Development: Build, iterate, and maintain AI agents within the architecture defined by the Lead AI Engineer. Own your agents end-to-end: prompt design, tool wiring, context routing, failure handling, and output validation.
• Data Integration: Integrate data across media platforms — ingesting, normalizing to schema, and routing to the correct agent context. Work within defined data contracts and surface schema drift before it becomes a runtime failure.
• Evaluation-First Development: No feature enters development without defined success criteria and regression tests. Run prompt benchmarking, track output quality across model versions, and proactively flag hallucination patterns or regressions. Evaluation is not a post-build step.
• Pipeline & ETL Work: Build and maintain ETL/ELT pipelines supporting daily automated callouts and weekly optimization reporting. Own data freshness and pipeline reliability.
• MCP Connector Work: Extend and operate MCP connectors for external platform APIs. Handle rate limits, retries, and failure modes — connectors must be production-ready.
• Human-in-the-Loop Workflows: Build and maintain Slack-based approval flows, feedback capture, alerts, and operational notifications — the key interface between AI and human decision-makers.
• Production Reliability: Own the reliability of your agents in production. Monitor outputs, respond to incidents, and drive root-cause fixes. Escalate risks early to the Lead AI Engineer.

CANDIDATE PROFILE

• 4+ years in software engineering, data engineering, ML, or AI platform roles with ownership of production systems
• Experience with media platform APIs (Google Ads, Meta, DV360, Semrush, SerpAPI)
• Strong Python and SQL (production-level)
• Experience with MCP or similar integration layers
• Hands-on work with LLM applications, agent systems, or tool-calling workflows
• Workflow orchestration tools: Airflow, Dagster, Prefect, dbt
• ETL/ELT pipelines and data reliability (schema management, contracts, monitoring)
• Cloud: AWS, GCP, or Azure; containerization (Docker)
• Experience with evaluation frameworks and model quality metrics
• Slack API and webhook-based automation
• Familiarity with vector databases and RAG patterns
• Experience combining LLM logic, business rules, and human approval flows
• LLM evaluation tooling (cost tracking, hallucination detection, benchmarking)

THIS ROLE IS NOT

• Generalist development without LLM or agent systems experience
• Prompt engineering without responsibility for data quality and pipelines
• Low-code automation (n8n, Make, Zapier)
• Roles with loose evaluation standards
• Research-only AI roles without production ownership
• Environments where shipping late is acceptable
• Roles without accountability for production incidents
• Positions requiring full architectural guidance before starting

WHAT WE OFFER

• Hybrid model: 3 days in the office (Tuesday–Thursday)
• Competitive salary and growth opportunities
• Private medical care (Medicover)
• Multisport card
• Annual education budget: $250
• Employee referral program
• Catered lunch every Tuesday
• Snacks and occasional breakfasts in the office