Note: The job is a remote job and is open to candidates in USA. Indotronix Avani Group is seeking a Lead AI Engineer for a 6+ months contract role based in St Petersburg, FL with 100% remote work. The role involves leading engineering efforts in building enterprise-grade AI software, particularly focusing on large language models and agentic systems. Responsibilities include developing AI coding tools, integrating AI systems, ensuring security in regulated environments, and providing technical leadership to engineering teams.
Responsibilities
- 8+ years of senior engineering experience building enterprise-grade software that other engineers or business teams depend on — internal platforms, large-scale services, developer tooling, SDKs, or similar systems
- At least two of those years must be deeply hands-on with large language models or agentic systems (this is non-negotiable for this role)
- Prior experience building or extending agentic coding tools like Claude Code or OpenAI Codex, and IDEs like VS Code — or comparable tools
- Strong Python skills, and enough Node.js / TypeScript to be effective on the parts of the platform that require it
- Direct experience designing or substantially extending integrations between AI coding tools and external systems (any of: Model Context Protocol servers, custom LangChain/LangGraph tools, Copilot extensions, JetBrains AI plugins, Cursor / Continue / Aider integrations)
- A working point of view on at least one production AI coding tool — enough to reason intelligently about prompt design, tool-use loops, context window economics, and where these systems tend to fail
- Experience with more than one LLM provider, and a clear position on building provider-agnostic abstractions so the platform isn't trapped on a single vendor's roadmap
- Cross-platform engineering experience across Windows, macOS, and Linux, including the realities of installers, path handling, shells, and credential storage on each
- Background in one or more of the technology stacks our application teams use (Python, Java/Spring, Angular, React, .NET, Oracle, Redshift, SQL Server) — enough to be credible reviewing stack-specific contributions
- A track record of shipping infrastructure that other engineers depend on every day, and being accountable when it breaks
- Comfort working in a regulated environment — financial services — with the discipline that implies around secrets, data classification, change management, and auditability, and practical experience with the security and data-protection concerns specific to LLM systems: prompt-injection mitigation, secret-leak prevention through prompts and tool outputs, PII redaction, and trust boundaries around model output
- Deep, hands-on knowledge of the software development lifecycle and modern engineering methodologies — agile, trunk-based development, code review, CI/CD, and observability-driven operations — and experience operating in a mature CI/CD and observability environment (e.g., Azure DevOps, Jenkins, SonarQube, and a log or APM platform such as Splunk, Datadog, or Dynatrace)
- Demonstrated technical leadership without direct authority — you have written architecture decision records that drove real consensus, presented technical direction to senior engineering and business leadership, and mentored other senior engineers (not just juniors) into stronger work
- Prior work on agent evaluation — golden task sets, regression suites, judge models, or other ways of measuring whether an AI workflow is actually getting better over time
- Active engagement with the developer-tools, AI agents, or AI/ML open-source community — substantive contributions, conference talks, or published writing that demonstrates depth and a public point of view
Skills
- 8+ years of senior engineering experience building enterprise-grade software that other engineers or business teams depend on — internal platforms, large-scale services, developer tooling, SDKs, or similar systems. At least two of those years must be deeply hands-on with large language models or agentic systems (this is non-negotiable for this role)
- Prior experience building or extending agentic coding tools like Claude Code or OpenAI Codex, and IDEs like VS Code — or comparable tools
- Strong Python skills, and enough Node.js / TypeScript to be effective on the parts of the platform that require it
- Direct experience designing or substantially extending integrations between AI coding tools and external systems (any of: Model Context Protocol servers, custom LangChain/LangGraph tools, Copilot extensions, JetBrains AI plugins, Cursor / Continue / Aider integrations)
- A working point of view on at least one production AI coding tool — enough to reason intelligently about prompt design, tool-use loops, context window economics, and where these systems tend to fail
- Experience with more than one LLM provider, and a clear position on building provider-agnostic abstractions so the platform isn't trapped on a single vendor's roadmap
- Cross-platform engineering experience across Windows, macOS, and Linux, including the realities of installers, path handling, shells, and credential storage on each
- Background in one or more of the technology stacks our application teams use (Python, Java/Spring, Angular, React, .NET, Oracle, Redshift, SQL Server) — enough to be credible reviewing stack-specific contributions
- A track record of shipping infrastructure that other engineers depend on every day, and being accountable when it breaks
- Comfort working in a regulated environment — financial services — with the discipline that implies around secrets, data classification, change management, and auditability, and practical experience with the security and data-protection concerns specific to LLM systems: prompt-injection mitigation, secret-leak prevention through prompts and tool outputs, PII redaction, and trust boundaries around model output
- Deep, hands-on knowledge of the software development lifecycle and modern engineering methodologies — agile, trunk-based development, code review, CI/CD, and observability-driven operations — and experience operating in a mature CI/CD and observability environment (e.g., Azure DevOps, Jenkins, SonarQube, and a log or APM platform such as Splunk, Datadog, or Dynatrace)
- Demonstrated technical leadership without direct authority — you have written architecture decision records that drove real consensus, presented technical direction to senior engineering and business leadership, and mentored other senior engineers (not just juniors) into stronger work
- Prior work on agent evaluation — golden task sets, regression suites, judge models, or other ways of measuring whether an AI workflow is actually getting better over time
- Active engagement with the developer-tools, AI agents, or AI/ML open-source community — substantive contributions, conference talks, or published writing that demonstrates depth and a public point of view
Company Overview