Senior AI Engineer (Internal Tools & Agents)

Trust Wallet · Remote - Global · Engineering

Posted 2026-10-02

Apply for this role →

About the Role

We are looking for a Senior AI Engineer to build AI-powered tools and Agent workflows that help our internal teams work more effectively.

Reporting to the Lead AI Engineer, you will turn operational needs and repetitive workflows into reliable AI products. You will work across the full development lifecycle—from understanding the problem and designing the workflow to building the backend, integrating internal systems, evaluating quality, and improving the tool after launch.

This is a hands-on engineering role for someone who has shipped LLM or Agent applications into production and understands that a useful internal tool must be accurate, secure, observable, and genuinely adopted by its users.

Responsibilities

Build internal AI tools and Agent workflows

- Design, build, and maintain AI-powered tools that improve internal workflows and decision-making.

- Develop Agent workflows involving context management, tool calling, knowledge retrieval, structured outputs, state management, and human approval or escalation.

- Integrate AI systems with internal data sources, APIs, business systems, and third-party services.

- Build backend services and APIs that support reliable and scalable AI applications.

Deliver measurable business impact

- Work directly with internal stakeholders to understand their workflows, pain points, and success criteria.

- Translate loosely defined business problems into clear, testable, and deliverable technical solutions.

- Measure adoption, task-completion quality, time saved, error reduction, and other relevant indicators after launch.

- Use real user feedback and production failures to prioritize improvements.

Build evaluation and quality systems

- Create evaluation datasets and regression tests for prompts, models, retrieval pipelines, tools, and end-to-end workflows.

- Establish release checks to identify quality regressions and unintended side effects before deployment.

- Compare models and implementation approaches based on quality, latency, reliability, security, and cost.

- Turn production failures and user feedback into reusable evaluation cases.

Ensure reliability, security, and observability

- Implement appropriate handling for timeouts, retries, rate limits, asynchronous jobs, partial failures, and third-party dependency issues.

- Build logging, tracing, monitoring, versioning, and recovery mechanisms for AI workflows.

- Apply appropriate access controls, data-handling rules, human approval steps, and audit trails for sensitive internal workflows.

- Diagnose whether failures originate from model behavior, missing knowledge, retrieval quality, tool responses, business rules, or backend services.

Collaborate and communicate proactively

- Work closely with the AI Engineer team and stakeholders across Product, Engineering, Operations, and other functions.

- Provide clear recommendations on feasibility, trade-offs, risks, cost, and alternative approaches.

- Break larger initiatives into practical, independently valuable delivery stages.

- Communicate progress, blockers, expected timelines, and next steps without requiring repeated follow-up.

- Document systems and decisions so that tools can be maintained and extended by others.

Requirements

- At least 4 years of software engineering experience, with strong backend engineering fundamentals.

- At least 1–2 years of hands-on experience building LLM-powered applications, Agent workflows, or AI automation systems.

- Experience shipping AI applications to real users and improving them based on production data and feedback.

- Practical experience with tool calling, RAG or knowledge retrieval, workflow orchestration, structured outputs, and stateful Agent systems.

- Experience building production services, APIs, data pipelines, and integrations with internal or third-party systems.

- Strong hands-on Golang proficiency. Golang is used for production development, and candidates must be able to complete the coding assessment in Golang.

- Practical Python experience for LLM and AI application development.

- Experience creating evaluation datasets, regression tests, monitoring, and release validation for LLM applications.

- Strong debugging skills across models, prompts, retrieval systems, tools, application logic, and backend infrastructure.

- A strong ownership mindset: you surface risks early, follow through after launch, and verify that what you built delivers the intended outcome.

- Fluent spoken and written English. English is the working language for daily communication and collaboration.

- Clear communication skills in a distributed, cross-functional environment.

Nice to Have

- Experience building internal productivity tools, enterprise AI applications, or workflow automation platforms.

- Experience with MCP, Agent frameworks, LLM gateways, tracing platforms, or model evaluation tooling.

- Experience with permission systems, audit logs, sensitive-data handling, or human-in-the-loop workflows.

- Experience working in crypto, Web3, financial services, or another security-sensitive environment.

- Familiarity with self-custody wallets, blockchain data, on-chain transactions, or crypto operations.

- Experience deploying and operating services using cloud infrastructure, containers, and Kubernetes.

Apply for this role →

← Back to all jobs