With capabilities spanning digital twins, normal behavior modeling, natural language processing, and computer vision, Avathon delivers real-time predictive intelligence and agentic decision-making at industrial scale.
Learn more at: avathon.com
Role Overview
The Forward Deployed AI Scientist is the AI and modeling owner for Avathon's customer deployments. Our applications in planning, logistics, PLM, asset performance, and machine vision all run on the same platform core: a Failure Prediction and Remaining Useful Life (RUL) model, a Computational Knowledge Graph that captures ontology and links industrial data, a Model Hub that serves ML and LLM models, and AI agents that reason over all of it. You make that core work on each customer's data — adapting models, mapping data into the graph, and proving results in live operations.
You will work side by side with customer engineers, operators, and Avathon's forward deployed engineers and design leads — taking models from customer data to production and bringing what you learn in the field back to the Platform AI team, so every application gets better at once.
Key Responsibilities
Core Models in Customer Environments
Adapt and tune Failure Prediction and Remaining Useful Life (RUL) models and anomaly detection for each customer's time series and event data — setting baselines, thresholds, and alert logic that operators trust
Build data quality and gap-filling methods for messy industrial sensor and business data at client sites, and validate signal quality before models go live
Evaluate and adapt time series and multimodal foundation models against customer data and use cases, and recommend what to deploy
Test and benchmark models on customer history to set clear accuracy, false-alarm, and value targets for each deployment
Knowledge Graph & Agents
Map raw customer data into the Computational Knowledge Graph — leading ontology discovery and entity resolution across historians, ERP, CMMS/EAM, PLM, and document sources
Evaluate and configure AI agents that query the graph, call models, and recommend actions in the customer's environment — with clear guardrails and human approval
Design customer-specific evaluation sets and acceptance criteria so the team and the client know when an agent or model is good enough to ship
Deployment & Productization
Work with the Model Hub team and FDE engineers on model serving, monitoring, retraining, and cost for each deployment
Turn repeated customer work into reusable platform services, templates, and configuration patterns that application teams can use without starting from scratch
Feed field learnings, model gaps, and new requirements back to the Platform AI and product teams
Client Engagement & Communication
Lead technical working sessions with customer data science, engineering, and operations teams — scoping use cases, data sources, and success metrics
Explain model behavior, results, and limitations clearly to operators, customer leadership, and Avathon's engagement directors
Write up methods and results — model cards, evaluation reports, and deployment notes — for product, customers, and leadership
Identify opportunities to extend AI capabilities within accounts — new assets, sites, data sources, or use cases — and surface them to the engagement director
Preferred Qualifications
Strong foundation in machine learning, with depth in at least two of: time series and anomaly detection, knowledge graphs and graph learning, LLMs and agents
Experience building ML or AI capabilities used across many products or customers
Experience deploying models with or for customers — in a forward deployed, solutions, or customer-facing applied science role
Experience at Palantir, C3 AI, Databricks, Google, AWS, Microsoft, or an industrial AI company is a strong plus
Familiarity with industrial data sources: historians (AVEVA PI, Aspen IP.21), IoT/SCADA, ERP, CMMS/EAM, and PLM systems
Ability to explain complex models clearly to non-technical audiences and build trust with operators and customer leadership
Should have client-facing experience in Industrial AI settings
Publications or patents in related areas are a plus
Qualifications
PhD in Computer Science, Machine Learning, Statistics, or a related field, or an MS with 5+ years of relevant industry experience
Strong Python and the standard ML stack (PyTorch, Hugging Face, LLM frameworks)
Experience shipping models to production, not only notebooks
Willingness and ability to travel up to 50%; based in Pleasanton, CA
Must be authorized to work in the United States without current or future company sponsorship
Compensation & Benefits
California Base Salary: $160,000 - $240,000 Pay is based on a number of factors including geographic location, job-related knowledge, skills, and experience
Equity in the form of stock options
Comprehensive health, dental, and vision coverage
401(k) participation
Generous PTO and company holidays
All travel expenses covered per Avathon policy
Avathon is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, pregnancy, genetic information, disability, status as a protected veteran, or any other protected category under applicable federal, state, and local laws.
Avathon is committed to providing reasonable accommodations throughout the recruiting process. If you need a reasonable accommodation, please contact us to discuss how we can assist you.