S

Sparta Commodities

Cloud-based commodities trading analytics platform

Staff AI Engineer - Agents & Harness

AI Application EngineeringPosted on Nov 17
Full-timeRemote

About the Job

Requirements

  • At least 8 years of experience building production software, including designing and shipping LLM-powered agents or agentic products used by real users.
  • Experience setting architecture across a team, leading designs through review, and mentoring others.
  • Hands-on knowledge of agent systems, including tool calling, multi-agent orchestration, streaming, context and memory management, retrieval, and failure modes over long runs.
  • Ability to explain trade-offs to engineering, product, and executive audiences and influence decisions with evidence.

Responsibilities

  • Author the target architecture for the agent platform, including the execution graph, LLM interfaces, tool contracts, memory and context management, sub-agents, sandboxing, and how skills plug in; write designs, run reviews, and own trade-offs.
  • Develop context engineering across branched conversations, handoffs between specialist agents, a decision-model layer that allows models to change without modifying agent code, and resumable, observable runs.
  • Define agent evaluation using golden question sets, offline and online evaluations in CI, tracing and feedback loops, and release gates for model or prompt changes.
  • Design the agent-to-agent API, MCP server and headless modes, and boundaries for what is exposed, what customers connect, and what stays internal.
  • Own agent reliability, cost, and safety, including latency and token budgets, failure handling, guardrails, provenance for every number, and the security model for agents that call tools and run code.
  • Mentor engineers, coach the team through design and code review, and spread agent-building practices, including AI-assisted development, across engineering.
  • Line-manage a group of engineers on the Agents team; run 1:1s, set goals, give feedback, support career growth, and take part in hiring.
  • Work with the Team Lead, product, and customers to shape the roadmap, decide what to build, buy, or stop, and size the risk in new ideas through quick prototypes.

Desired Qualifications

  • Experience building an agent harness or framework, contributing to an open-source agent framework such as OpenAI Agents SDK, Claude Agent SDK, LangGraph, or DSPy, or building MCP servers or clients.
  • Knowledge of knowledge graphs, ontologies, or entity resolution for mapping trader language to relevant data.
  • Experience with sandboxed code execution, guardrails, or security work for agents.
  • Fine-tuning or model-selection experience, or a research background.
  • Interest in commodities or financial markets.

About the Company

Sparta Commodities

Cloud-based commodities trading analytics platform

Sparta Commodities provides a cloud platform for commodity trading that delivers real-time pricing, arbitrage data, and shipping costs. Users access Sparta Cloud to run price blends, view live feeds, and calculate arbitrage and freight costs, with automated workflows and alerts that notify them of opportunities. The platform stands out by offering end-to-end automation and real-time analytics across price discovery, arbitrage, and freight data within a single subscription-based toolset. Its goal is to boost trader productivity and speed by providing timely data and automated processes so traders can act on opportunities faster.

Company size
201-500
Company stage
Series B
Total funding
$65.5M
Headquarters
Geneva, Switzerland
Founded
2020

Recently Posted Jobs

Explore more opportunities at Sparta Commodities

See more jobs