[Remote] Machine Learning Engineering Lead

Note: The job is a remote job and is open to candidates in USA. Sightly is a growing technology company leading the revolution in real-time marketing and brand intelligence. They are seeking a Machine Learning Engineering Lead to drive the development of enrichment models, optimize ad systems, and provide technical leadership for a small team in a fully remote environment.

Responsibilities

• Enrichment models across our cultural data pipeline: entity extraction, topic and stance classification, embeddings, clustering, sentiment, brand safety, and related tasks across billions of news and social records
• Multi-modal enrichment for image and video signals from social platforms, complementing our text-heavy core
• Ad optimization systems built from the ground up, including bid optimization, budget allocation, creative selection, audience targeting, or related problems, grounded in historical performance data and well-reasoned heuristics
• Experimentation design and execution: framing the question, choosing the right test, instrumenting it, and producing results the business can act on
• Production ML infrastructure on GCP: training, evaluation, deployment, monitoring, and the glue that keeps models reliable as data shifts
• Technical leadership for a small ML team, including code review, mentorship, prioritization, and raising the bar on rigor without slowing delivery
• Cross-functional partnership with Data Engineering on pipeline integration, and with Account Management and Performance Managers to translate business problems into model problems

Skills

• Min. Experience: Senior Level
• Strong foundation across classical ML, neural networks, and Transformers, reaching for the right tool rather than the trendiest one
• Comfortable with both supervised and unsupervised paradigms: classification, regression, clustering, dimensionality reduction, representation learning
• Practical fluency with NLP and at least working familiarity with computer vision for image and video enrichment
• Understanding of when a simple model beats a complex one, and the discipline to ship the simple one
• Track record of structuring and running experiments end-to-end: hypothesis, design, instrumentation, analysis, decision
• Comfortable with ad hoc statistical testing, picking the right test for the task, reasoning about power, controlling for confounds
• Knows the difference between a model that benchmarks well offline and one that holds up in production
• Research mindset paired with a shipping mindset: rigorous, but allergic to research-for-its-own-sake
• Experience building optimization systems, whether mathematical optimization, heuristics, or learned policies, applied to a real-world domain
• Comfortable reasoning about objective functions, constraints, and tradeoffs in messy business contexts
• Strong Python and the standard ML stack: scikit-learn, PyTorch, TensorFlow, HuggingFace, NumPy, pandas
• FastAPI and async/await patterns for serving models and building ML-facing services
• Experience working with data at scale, including the practical realities of billions of records: partitioning, sampling, distributed processing, cost management
• GCP for training, serving, and infrastructure, such as Vertex AI, Cloud Run, GCS, or equivalent
• PostgreSQL and Snowflake for working with large-scale data
• Docker and CI/CD pipelines for reproducible, deployable ML workloads
• Comfortable with the realities of production ML: data drift, retraining cadence, monitoring, cost management
• Experience leading or mentoring engineers, even informally, through code review, technical direction, and raising the bar on quality
• Strong collaboration habits with Data Engineering, and the ability to translate fluently between technical and business audiences
• Can sit with an Account Manager or Performance Manager, understand what they actually need, and turn it into a tractable modeling problem
• Clean code habits, sensible architecture, strong typing discipline
• Test-driven mindset for ML code: covering data assumptions, edge cases, and regression paths, not just happy paths
• Comfortable with modern dev practices: Git, code review, CI/CD
• Advertising, adtech, or media industry experience
• Familiarity with LLMs and modern AI tooling, useful context for the broader engineering org but not the focus of this role
• Causal inference or uplift modeling background
• Experience with recommendation systems or ranking
• 5+ years of ML experience, ideally with a foundation built before the LLM era

Company Overview

Company H1B Sponsorship