Engineering Lead (AI & Automation Products)

Engineering Lead (AI & Automation Products) is a full-time, permanent position with Dentsu Global Services Private Limited, under the Merkle brand, in Bengaluru, India. The role requires working hours that overlap with US Eastern Time and collaboration across US and India teams.

Required qualifications and work:

• Bring 12–16 years of professional software engineering experience, deep Python expertise, and demonstrated leadership or management of engineers through code review, mentoring, and growth planning.
• Design and build production Python APIs using FastAPI, Flask, or similar tools, and full-stack applications with React and Tailwind CSS. Apply relational database skills in schema design, normalization, and query performance; PostgreSQL is preferred.
• Use Claude Code or similar AI-assisted development tools effectively, including agentic coding and context management, and establish team standards for their use. Integrate LLM APIs such as Claude or OpenAI into production systems, covering system prompts, structured outputs, multimodal inputs, function calling, tool schemas, argument validation, tool results, chained calls, and basic failure and retry handling.
• Apply context-window management, token budgeting, long-document strategies, and retrieval or context selection. Design guardrails against prompt injection, adversarial inputs, and risks from external briefs, trafficking sheets, platform exports, and other untrusted model-readable inputs.
• Integrate third-party APIs using OAuth; apply production credential security and secrets management, ideally with Azure Key Vault or an equivalent service. Use unit and integration tests, data validation, golden datasets, regression evaluations, structured-output checks, and observability. Understand human-in-the-loop adjudication, labeled examples, category-level accuracy, feedback loops, and quality gates.
• Work with cloud infrastructure, preferably Azure, and containers, CI/CD, managed identities, object storage, and background processing. Implement queues or scheduled jobs with retries, idempotency, status tracking, and operational monitoring. Communicate effectively across distributed teams.

Preferred qualifications

• Exposure to LangChain, LangGraph, CrewAI, Temporal, Azure Durable Functions, Celery/RQ, or similar workflow tools; model selection and cost optimization, including caching, batching, latency and cost tradeoffs, and usage forecasting.
• Background in media, advertising, or marketing technology data; familiarity with Unity Catalog, attribute-based access control, tag-driven policies, MCP workflows, and data flywheel concepts.
• Exposure to DV360 Structured Data Files, TTD API, or comparable adtech formats and APIs; open-source contributions or public full-stack or AI engineering projects.