Lead Engineer - Ad Tech

You have 8+ years of software development experience, with experience designing, building and operating complex distributed systems through at least one complete implementation lifecycle. 

You have strong backend engineering fundamentals and are comfortable designing scalable APIs, microservices, asynchronous systems and data-intensive applications. 

You are fluent in Java / Spring and microservices architecture, with experience building highly available production systems. 

You have experience working with databases including RDBMS and NoSQL technologies such as Cassandra and MongoDB, and understand data modeling and storage trade-offs. 

You have experience building distributed event-driven architectures using technologies such as Kafka. 

You understand Ad Tech business fundamentals and how technology supports business objectives, and can translate business vision into technical strategy while understanding architectural and financial trade-offs. 

Experience building or integrating DSP, SSP or Ad Server technology platforms in support of self-service advertising is preferred. 

Applied AI & LLM Engineering 

You have hands-on experience designing, building and operating production applications using LLMs, beyond chat interfaces, prompt experimentation and proof-of-concept applications. 

You have designed and optimized production-grade RAG systems, with strong understanding of data ingestion and chunking, retrieval and ranking, context and grounding, hybrid retrieval, and evaluation. 

You have experience with knowledge graphs and graph-based retrieval, and understand when to use vector search, Graph RAG, structured queries, traditional search or hybrid approaches based on the problem and data. 

You understand the engineering trade-offs of production LLM systems, including quality, latency, cost, reliability, observability, security and failure handling, and can systematically improve retrieval and overall system performance rather than relying primarily on prompt engineering. 

Agentic Systems 

You have designed or built agentic applications or orchestration frameworks where LLM-powered components interact with tools, APIs, retrieval systems and other agents to accomplish multi-step tasks. 

You understand concepts such as: 

• Tool/function calling 

• Agent planning and execution 

• Workflow and state management 

• Multi-agent orchestration 

• Agent memory and context management 

• Routing and delegation 

• Human-in-the-loop workflows 

• Guardrails and policy enforcement 

• Retry, timeout and fallback strategies 

• Agent observability and traceability 

• Evaluation of agent behavior and task completion 

You understand the distinction between deterministic and non-deterministic execution paths and can design systems that deliberately combine both. 

You know when a business workflow should remain deterministic and testable using conventional software and when probabilistic reasoning using an LLM or agent is appropriate. 

You design AI systems assuming that model outputs can be incorrect or unpredictable and therefore incorporate appropriate validation, constraints, structured outputs, fallbacks, idempotency, observability and human intervention where required. 

AI Quality & Evaluation 

You understand that production AI systems require rigorous quality and evaluation practices beyond model selection and prompting. 

You have experience establishing offline and online evaluation strategies for LLM-powered systems, covering retrieval and response quality, factual grounding, task completion, reliability, latency and cost. 

You can build evaluation harnesses and regression tests with representative datasets, metrics and quality thresholds, and integrate them into CI/CD as deployment and quality gates. 

You can use evaluation, observability and tracing to diagnose quality issues across data, retrieval, context, prompts, models and orchestration, and drive systematic improvements. 

Engineering Leadership 

You can translate ambiguous business problems into clear technical architectures and incrementally deliver solutions from experimentation through production. 

You make architecture decisions based on measurable trade-offs rather than technology trends and are comfortable challenging unnecessary complexity. 

You can lead POCs and technical experiments while clearly distinguishing between what demonstrates feasibility and what is required to operate the capability reliably at enterprise scale. 

You have proven technical leadership capabilities and the ability to influence engineers, product leaders and cross-functional stakeholders.