Palantir AI Engineer

Trusty Bytes emphasizes clarity, reliability, and speed, with success measured by business outcomes such as stability, performance, and cost efficiency. A people-centric culture, founder-led engagement, and a strong focus on both customer and employee experience define how the organization operates.\n\nExperience: 4 to 8 years (Relevant 3 + in Palantir)\nWork Location: Gurgaon (Work from client location)\n\nRole Overview:\nAs an AI Engineer, you will be the architect of our intelligence layer. You won't just be building models in isolation; you will be leveraging Palantir Foundry and AIP to create integrated, "human-in-the-loop" AI applications.

Your goal is to transform static data assets into dynamic, agentic workflows that solve complex operational challenges.\n\nKey Responsibilities\n* AIP Logic & Tooling: Design and implement AIP Logic functions and integrate LLMs (Large Language Models) into Foundry workflows.\n* Ontology Architecture: Map complex business processes into the Foundry Ontology, ensuring AI models have a structured, semantic understanding of the organization.\n* Pipeline Development: Build robust data pipelines using PySpark and Foundry Data Connection to feed high-quality data into AI features.\n* Prompt Engineering & Orchestration: Develop and refine prompts and use AIP Assist or Automate to create autonomous agents and decision-support tools.\n* Application Building: Use Workshop and Slate to build intuitive front-end interfaces where end-users can interact with AI insights.\n* Governance & Ethics: Implement rigorous testing, versioning, and "checks and balances" within the AIP framework to ensure AI outputs are safe, reliable, and explainable.\n\nTechnical Qualifications\n- Palantir Core | Proficiency in Foundry Ontology, Workshop, Contour, and Pipeline Builder.\n- AI/ML Skills | Deep understanding of AIP Logic, LLM orchestration (RAG), and prompt engineering.\n- Languages | Expert-level Python and PySpark. Familiarity with TypeScript/SQL is a plus.\n- Data Engineering | Experience with data modeling, ETL/ELT, and managing large-scale datasets.\n- DevOps/Mojo | Knowledge of Foundry’s modeling objective (Mojo) and CI/CD for ML models.\n\nPreferred Experience\n* Experience building Agentic Workflows (AI that can take actions, not just answer questions).\n* A background in a highly regulated industry (Finance, Healthcare, Defense) where data privacy and security are paramount.\n* Palantir certifications (e.g., Foundry Data Engineer or Foundry Public Sector).\n\nThe Ideal Candidate\nYou are a "full-stack" data thinker. You don't wait for a clean dataset; you go into the Foundry filesystem, find what you need, model it into the Ontology, and wrap an AIP-driven solution around it.

You value security-first AI and understand that an LLM is only as good as the data it can access.