- Company Name
- X, The Moonshot Factory
- Job Title
- Machine Learning Engineer, AI Early Stage Project
- Job Description
-
**Job Title:** Machine Learning Engineer, AI Early Stage Project
**Role Summary:**
Design, develop, and deploy advanced machine learning systems that transform unstructured industrial data (P&ID diagrams, manuals, sensor streams, video feeds) into structured, queryable Process Knowledge Graphs (PKGs). Architect agentic Retrieval-Augmented Generation (RAG) workflows combining Vision‑Language Models (VLMs) and Large Language Models (LLMs) to create digital twins, automate continuous process optimization, and bridge perception, sensing, and graph‑based reasoning.
**Expectations:**
- Deliver production‑grade ML models within iterative agile sprints.
- Ensure high‑quality code, robust data pipelines, and scalable deployment.
- Validate model outputs against engineering requirements and iterate quickly.
- Communicate complex technical concepts to multidisciplinary teams.
**Key Responsibilities:**
1. Build and maintain multimodal data ingestion pipelines processing documents, images, and telemetry at scale.
2. Implement graph extraction and enrichment techniques to populate and refine PKGs from raw industrial artifacts.
3. Design and deploy agentic RAG frameworks where LLMs reason over knowledge graphs to generate actionable digital twins.
4. Engineer LLM‑driven code generation pipelines (function calling, tool‑use) that produce executable logic in Python, SQL, or Cypher.
5. Integrate computer vision models for object detection/segmentation on technical imagery and diagrams.
6. Apply MLOps best practices: model versioning, monitoring, A/B testing, and performance optimization.
7. Collaborate with research and domain experts to iterate from prototype to production‑ready systems.
8. Diagnose and mitigate high‑noise data challenges, ensuring resiliency and data‑quality standards.
**Required Skills:**
- Proficiency in Python, PyTorch (or JAX), and related ML libraries.
- Experience with LLMs (GPT, LLaMA, etc.) and VLMs (CLIP, BLIP, etc.) in applied settings.
- Strong grasp of graph data structures, Knowledge Graphs, and Graph Neural Networks; familiarity with Neo4j, NetworkX.
- Demonstrated ability in prompt engineering, fine‑tuning, and RAG implementation.
- Familiarity with LLM‑driven code generation, function calling, or tool‑use patterns.
- Solid background in computer vision for technical images, including object detection and segmentation.
- Understanding of MLOps pipelines, model deployment, monitoring, and continuous integration.
- Experience with agentic workflows (LangChain, AutoGen) and iterative reasoning loops.
- Knowledge of Reinforcement Learning concepts as applied to LLM fine‑tuning or optimization is a plus.
**Required Education & Certifications:**
- Bachelor’s degree in Computer Science, Artificial Intelligence, Computer Engineering, or equivalent practical experience.
- Minimum 3 years of software engineering and applied machine learning experience.
Mountain view, United states
On site
Junior
21-12-2025