Engineering · Remote

Principal Engineer, LLM

Upstart · Remote · Posted Nov 19, 2026

This posting comes from Upstart. Confirm the details on their site before applying.

Overview

At Upstart, our Applied AI team's mission is threefold: Agent Platform: Build and support the platform that makes it straightforward for non technical users across Upstart to build, launch, and operate AI agents (voice/chat/text) reliably. Borrower and Applicant Assistant Orchestration: Build the orchestration layer behind a unified personal assistant that shares context across specialized agents, giving applicants and borrowers a cohesive, assistant-like experience.

Requirements

  • Software engineering foundation and live AI shipping: 8+ years of software engineering experience, along with hands-on experience shipping non-deterministic, user-facing LLM or AI products to real users.
  • Agentic and LLM systems architecture: Deep practical experience designing production LLM systems involving structured outputs, tool calling, context engineering, retrieval/memory patterns, and multi-step or multi-agent orchestration (including state management and context sharing across agents).
  • Production systems, latency, and reliability engineering: Strong distributed-systems and backend fundamentals for non-deterministic pipelines: async orchestration, streaming UX, caching, timeouts and fallback paths, rate-limit handling, and end-to-end observability/tracing to diagnose failures.
  • Evaluation and quality engineering: Proven experience building golden datasets, automated evaluation harnesses (deterministic checks, heuristics, and LLM-as-a-judge), regression gates, and systematic failure-mode analyses.
  • Safety, guardrails, and applied AI pragmatism: Strong instinct to reach for the simplest reliable tool first (deterministic logic, prompting, context engineering, guardrails against hallucinations and prompt/data risks), paired with sound judgment on when latency, unit cost, or reliability justify…
  • Full-stack prototyping fluency: Comfortable moving fast across backend and frontend stacks (such as Python, TypeScript/React, and modern REST or streaming APIs) to put working end-to-end prototypes in front of users and stakeholders quickly.
  • Low-ego pathfinder collaboration and UX empathy: Track record aligning early with partner squads, handing off ownership cleanly without ego, and partnering directly with Product, Design, and operations or business leaders to turn ambiguous goals into concrete latency budgets and resilient fallback…
  • Bachelor's degree in Computer Science, Engineering, Mathematics, or equivalent practical experience.

Preferred qualifications

  • Startup or 0-to-1 incubation DNA: Experience operating in early-stage or low-boundary environments where you owned ambiguous problems end to end under real constraints.
  • Agent platform and shared abstractions: Experience designing reusable agent primitives, SDKs, or orchestration platforms that help peer engineering teams build, evaluate, and launch AI workflows safely and quickly.
  • Production fine-tuning and distillation reps: Hands-on experience fine-tuning or distilling smaller models in production to improve latency, unit economics, or task reliability.
  • AI-native engineering habits: Deep daily fluency with agentic coding workflows and automated verification harnesses, plus experience helping peer engineers adopt higher-leverage development practices.
  • Experience in regulated or high-trust consumer domains (such as fintech, lending, or healthcare) where accuracy, explainability, PII protection, and user trust are critical.

Benefits

  • Health insurance, Dental insurance, Vision insurance, 401(k), 401(k) matching, Paid time off, Paid sick leave, Parental leave, Life insurance, Disability insurance, HSA/FSA, Stock / equity, Bonus, Wellness program, Employee assistance program

Great fits start here.

Whether you're looking for your next role or your next team member, give people and possibilities a place to meet.