⚙️ Senior Machine Learning Engineer – Applied AI / Agent Systems

Company: AdsGency AI

📍 Location: Onsite (San Francisco City)

💼 Employment Type: Full-Time

🚚 Relocation to San Francisco City Required

🛂 We Sponsor OPT / CPT / STEM-OPT / DO NOT sponsor H1B Transfer

We’re AdsGency AI — an AI-native startup building a multi-agent automation layer for digital advertising.

Our system uses LLM and ML-driven agents to autonomously launch, scale, and optimize ad campaigns across Google, Meta, TikTok, and more — no human marketer required.

Our mission: build the operating system where AI runs performance marketing better than humans ever could.

We’re backed by top-tier investors and moving fast. This is your chance to join early — and help design the ML foundation that powers the next evolution of ad intelligence.

As a Senior Machine Learning Engineer, you’ll design, train, and deploy AI models that drive AdsGency’s agent intelligence — from ad performance prediction to cross-channel optimization and creative generation.

You’ll bridge the gap between data science, engineering, and systems design, shaping the brain of our multi-agent OS.

This role sits at the core of AdsGency’s intelligence layer — where models don’t just predict, but act.

• 🧠 Agent Intelligence Models: Develop and fine-tune models that predict campaign performance, bid pacing, and creative success.

• 📊 Reinforcement & Decision Systems: Build RL and multi-objective optimization frameworks enabling agents to learn from feedback and improve autonomously.

• 🧬 LLM + ML Hybrid Systems: Integrate generative agents (OpenAI, Claude, LangGraph) with quantitative models for adaptive decision-making.

• ⚙️ Data Pipelines: Architect and maintain scalable feature pipelines and embeddings for multi-platform ad data.

• 🔍 Measurement & Attribution: Design models to unify performance signals across Google, Meta, TikTok, etc., handling delayed and biased feedback.

• 📈 Experimentation Frameworks: Develop A/B testing and counterfactual learning systems to validate model improvements.

• 🚀 ML Infrastructure: Own the training → evaluation → deployment lifecycle using modern MLOps practices (e.g., Weights & Biases, Airflow, Docker).

Modeling & ML: PyTorch, TensorFlow, Scikit-learn, XGBoost, LightGBM, Hugging Face, Transformers

Languages: Python, Go (for systems), SQL

Infra & MLOps: AWS/GCP, Docker, Kubernetes, Airflow, Weights & Biases, MLflow

Data Systems: Kafka, PostgreSQL, Redis, Supabase, Qdrant/Weaviate (vector DBs)

AI Layer: OpenAI, Claude, LangChain, LangGraph, CrewAI

✅ 4–8 years of experience in ML engineering or applied data science

✅ Strong foundation in ML algorithms, model lifecycle, and feature engineering

✅ Proficiency in Python and ML frameworks (PyTorch/TensorFlow)

✅ Experience building models that go into production, not just notebooks

✅ Understanding of distributed systems, data pipelines, and model serving

✅ Experience with A/B testing, reinforcement learning, or online learning

✅ Curiosity about how LLMs and agents can augment traditional ML systems

✅ Startup mindset — fast iteration, ownership, and bias for impact

✨ Experience in AdTech / MarTech, especially prediction, attribution, or bidding systems

🧠 Experience integrating LLMs with structured data pipelines

⚙️ Knowledge of reinforcement learning, causal inference, or bandit algorithms

🌱 Prior work in early-stage or high-growth startups

🎯 Strong sense of product impact — you ship models that move metrics

• Competitive salary + meaningful equity

• Core ownership in a fast-scaling AI company

• Work directly with founders and research engineers on frontier agentic systems

• Culture of speed, autonomy, and craftsmanship — no corporate bureaucracy

• Build systems that redefine how advertising learns and optimizes itself

• Visa sponsorship (OPT / CPT / STEM-OPT / no H1B Transfer)

Industry: AI & Software Development

Employment Type: Full-Time

Location: Onsite (San Francisco City)