Machine Learning (ml) Engineer

Vectara Vectara · Data AI · Remote · Platform Engineering and ML

Vectara is seeking an ML Engineer to design, prototype, research, and build AI systems for their enterprise RAG platform. The role involves training, evaluating, and deploying ML models in NLP, Information Retrieval, AI Agents, and LLMs, with a focus on improving agent quality, multilinguality, self-supervised learning, agentic behavior, and hallucination reduction. The ideal candidate has 5+ years of experience applying ML to real-world problems, strong Python skills, and domain knowledge in RAG, LLM, information retrieval, or Multimodal LLMs.

What you'd actually do

  1. Design, prototype, research and build AI systems for Vectara.
  2. Train, evaluate and deploy ML models in the domains of Natural Language Processing, Information Retrieval, AI Agents, Large Language Models (LLMs) and Multimodal Large Language Model (MMLLMs).
  3. Improve the quality of Vectara’s AI Agents and RAG-as-a-service platform, working on features like multilinguality, self-supervised learning, agentic behavior and hallucination reduction.
  4. Publish technical blogs, papers, and patents.

Skills

Required

  • Python
  • pandas
  • transformers
  • torch
  • RAG
  • LLM
  • information retrieval
  • Multimodal LLMs
  • deep learning concepts
  • Transformers
  • Retrieval-Augmented Generation (RAG)
  • mixture of experts (MoE)
  • training ML systems end-to-end

Nice to have

  • PhD in Computer Science/Engineering
  • Publications in prestigious venues such as ACL, NAACL, EMNLP, NeurIPS, ICML, ICLR
  • ML engineer in an early-stage, high growth environment
  • agent orchestration
  • tool use
  • evals
  • guardrails
  • llm_observability
  • fine_tuning
  • model_serving
  • recommender_systems
  • search_ranking
  • vision
  • multimodal
  • audio_speech
  • frontier_research
  • interpretability
  • synthetic_data
  • agent_research
  • rl_post_training
  • rlhf
  • reward_modeling
  • rl_robotics
  • embodied_ai

What the JD emphasized

  • Accuracy, Security, and Explainability
  • Accuracy, Security, and Explainability
  • retrieval, embedding, reranking
  • LLM trained for quality
  • Hallucination Mitigation
  • production ready
  • machine learning
  • RAG, LLM, information retrieval, Multimodal LLMs.
  • Python
  • data curation to evaluation and deployment

Other signals

  • Enterprise AI Agents
  • RAG Platform
  • Hallucination Mitigation
  • LLM trained for quality
  • production ready AI