Ford Motor Company je avtomobilski proizvajalec s sedežem v Združenih državah Amerike, katerega zgodovina sega v leto 1903. Poleg razvoja in proizvodnje vozil podjetje širi svoje področje delovanja na storitve mobilnosti, digitalne izkušnje in programsko opremo. Ford se posveča tudi trajnostnim prizadevanjem in okoljskim ciljem ter s tem povezuje promet, inovacije, vpliv na skupnost in odgovorne poslovne prakse.
AI/ML Architect at Ford Motor Company (Hybrid, India)
Build autonomous agents, multi-agent systems, and retrieval-augmented generation pipelines for enterprise use. Deploy production generative AI on Google Cloud using MLOps, containers, and vector databases.
Opis
- Develop and deploy autonomous and semi-autonomous AI agents for enterprise processes.
- Design multi-agent architectures with orchestration, task breakdown, planning loops, and inter-agent communication.
- Connect agents to enterprise platforms, REST and GraphQL APIs, databases, and Google Cloud services.
- Build short-term, long-term, episodic, and semantic memory capabilities.
- Create, test, refine, and maintain prompt templates and reusable prompt libraries.
- Assess, choose, and fine-tune foundation models, including Gemini through Vertex AI Model Garden.
- Build and tune high-throughput, low-latency RAG pipelines, including document ingestion, chunking, and embedding generation.
- Integrate and operate scalable vector databases.
- Clean and prepare structured and unstructured data for use.
- Provision secure Google Cloud environments with Terraform infrastructure as code.
- Build and maintain CI/CD pipelines for automated tests, container builds, and deployment across environments.
- Package applications and agents with Docker, then deploy them on GKE or Cloud Run.
- Set up workflow orchestration with tools such as Vertex AI Pipelines or Cloud Composer/Airflow.
- Define metrics and testing pipelines to assess agent quality, latency, token use, and hallucination rates.
- Add content filtering, moderation, grounding checks, prompt-injection defenses, privacy controls, and human review gates.
- Set up production monitoring, logs, distributed tracing, and real-time alerts.
- Improve agent workflows, model settings, caching, and infrastructure for cost and performance.
- Manage Git branches, pull requests, and code reviews.
- Work with data scientists, software engineers, and business teams to deliver production-ready AI solutions.
- Document system architecture, agent flows, tools, prompts, containers, and operating procedures.
Zahteve
- At least three years of professional experience in machine learning or AI engineering, with a solid MLOps foundation.
- One to two years of practical experience designing and deploying production LLM, generative AI, or agentic systems.
- Expert proficiency in Python.
- Experience with PyTorch, TensorFlow, or Scikit-learn.
- Strong theoretical and practical knowledge of deep learning, natural language processing, and Transformer architectures.
- Practical experience with LangChain, LangGraph, LlamaIndex, CrewAI, or Vertex AI Agent Builder.
- Demonstrated experience with Google Cloud Platform, Docker, Kubernetes/GKE, Cloud Run, and Cloud Functions.
- Demonstrated experience with vector indexing, semantic search, Vertex AI Vector Search, AlloyDB, or comparable tools.
- Knowledge of SQL and NoSQL databases, REST APIs, and GraphQL APIs.
- Experience with Terraform, Git, and CI/CD tools such as Cloud Build, GitLab CI, or Jenkins.
- Experience with Prometheus, Cloud Logging, or LangSmith, and with evaluating LLM output quality and safety.
- Google Cloud Professional Machine Learning Engineer or Professional Cloud Architect certification is highly desirable.
- Bachelor’s degree or equivalent.