UST
Website:
ust.com
Job details:
Role Description
Background
This role is for an
AI Engineer who operates an existing multi‑agent orchestration platform to build agentic applications and redesign business‑critical workflows. You will not be inventing the orchestration layer; instead, you will operate it as a
high‑leverage control plane—framing the right problems, designing effective agent workflows, and embedding AI‑driven solutions into real systems with real data, constraints, and controls.
We are looking for an individual who can operate like a
self‑contained force multiplier—able to take a domain, a workflow, and a set of tools and turn them into a durable, production‑ready agentic system. The best people in this role think like systems engineers, communicate like product managers, and execute with the ownership mindset of a tech lead.
This role operates in a
24×6 shift model, supporting continuous business operations.
What You’ll Do
- Work closely with stakeholders across trading, engineering, data, operations, risk/finance, and product to identify high‑value workflows, clarify constraints and SLAs, and define what success looks like.
- Design and implement agent workflows by decomposing complex, ambiguous problems into well‑scoped, executable tasks.
- Define agent roles such as planner, implementer, reviewer, analyst, and ops/SRE, along with clear prompts, acceptance criteria, and guardrails.
- Operate and refine agentic systems using tools such as Claude, ChatGPT, GitHub Copilot, and Cursor to design, review, refactor, and harden solutions.
- Treat workflows as systems rather than scripts—building in observability, failure handling, iteration, and feedback loops from day one.
- Review and harden AI‑generated outputs for correctness, safety, performance, and alignment with business constraints.
- Embed AI solutions into production systems, integrating with real data sources, APIs, security controls, and operational processes.
- Participate in and support a 24×6 operational model, ensuring reliability and continuity of AI‑driven workflows.
Qualifications
Must Have
- 3–6 years of software engineering experience building and operating backend or data‑intensive systems in production environments.
- Strong computer science fundamentals, including data structures, algorithms, complexity, distributed systems basics, concurrency, networking, and database fundamentals.
- Hands‑on experience using modern LLM/AI tools such as Claude, ChatGPT, GitHub Copilot, Cursor (or similar) to design, write, refactor, and review production code and workflows.
- Experience building internal tools or workflows that programmatically interact with LLMs, including areas such as tool/function calling, prompt engineering, RAG, automation, or AI‑assisted data workflows.
- Proficiency in at least one major programming language such as Python, TypeScript/Node.js, or Go.
- Familiarity with modern cloud and DevOps tooling (AWS, Azure, or GCP; containers; CI/CD pipelines; monitoring and logging).
Demonstrated Ability To
- Decompose ambiguous business problems into clear, agent‑executable tasks.
- Write crisp specifications, prompts, workflows, and acceptance criteria that agents and humans can reliably execute.
- Think in systems—anticipating failure modes, dependencies, and feedback loops across services, data, and teams.
- Review and harden AI‑generated work to ensure quality, safety, performance, and maintainability.
- Operate independently with high ownership, acting as a force multiplier across domains, workflows, and tools.
Skills
software engineering,python,typescript,large language model,go,node.js,
Click on Apply to know more.