Tech Holding
Tech Holding
Tech Holding is a technology services company offering professional services, managed services, and staffing solutions. It works with businesses on digital transformation and operational needs, drawing on industry knowledge and technical expertise to support growth and strategic objectives. Its staffing work connects clients with talent, alongside consulting and managed services for organizations navigating digital change.

Remote Contract Machine Learning Engineer

Build NLP and large language model systems for Tech Holding’s enterprise risk intelligence platform. Deploy monitored services that help detect disinformation, reputational attacks, and security threats.

Opis

  • Create and deploy machine learning models for text classification, entity extraction, clustering, and narrative and topic detection.
  • Develop LLM features using prompt design, retrieval-augmented generation, and agentic workflows.
  • Build embedding and semantic search pipelines to connect related content and reveal emerging threats.
  • Turn models into scalable, monitored services with defined latency, cost, and accuracy targets.
  • Create evaluation frameworks and datasets to track model quality, bias, and drift over time.
  • Partner with data engineering to improve ingestion, labeling, and feature pipelines for high-volume, noisy, multilingual text.
  • Monitor production models, retrain them, and investigate quality issues end to end.
  • Record technical approaches and explain trade-offs to technical and non-technical stakeholders.
  • Collaborate as an individual contributor with data engineers, backend engineers, and product stakeholders.
  • Contribute to an AI risk intelligence platform that helps enterprises detect, assess, and respond to disinformation, reputational attacks, and security risks.

Zahteve

  • At least four years of professional experience building and deploying production machine learning models.
  • Strong Python skills and practical experience with PyTorch, TensorFlow, or Hugging Face Transformers.
  • Applied NLP experience with real-world text, such as classification, named entity recognition, topic modeling, or semantic similarity.
  • Hands-on experience building with LLMs, including OpenAI, Anthropic, or open-source models, and using prompt engineering and RAG.
  • Experience with embeddings and vector search tools such as pgvector, Pinecone, OpenSearch, or FAISS.
  • Experience deploying and operating ML services on AWS, GCP, or Azure, with Docker and CI/CD.
  • Solid understanding of model evaluation, including metrics, test sets, error analysis, and production monitoring.
  • Strong communication skills and ability to work in a remote, cross-functional team.
  • Must be authorized to work for any U.S. employer; visa sponsorship is not available.
  • Preferred: Background in trust and safety, threat intelligence, disinformation research, or social media data.
  • Preferred: Experience building agentic AI systems or tool-using LLM workflows with LangChain, LlamaIndex, or custom frameworks.
  • Preferred: Experience with multilingual NLP.
  • Preferred: Familiarity with MLOps tools such as MLflow, SageMaker, Vertex AI, or Weights & Biases.
  • Preferred: Experience with streaming or large-scale data processing tools such as Kafka, Spark, or Airflow.
  • Preferred: Experience fine-tuning LLMs with LoRA or PEFT, or distilling models to reduce cost and latency.

Ugodnosti

  • Equal opportunity employer committed to a diverse and inclusive workplace.
  • Application accommodations are available upon request.

Sorodna delovna mesta