Principal AI Engineer

Principal AI Engineer

TechGrove AI Engineering Pod

You build the intelligent features that make software feel like magic, and you make them work in production. You are equally at home shipping a clean API, standing up a RAG pipeline, evaluating an agent, or tuning guardrails so an LLM behaves safely at scale. You are excited by the current moment in AI and pragmatic about where it actually adds value.

As Principal AI Engineer, you design and build the AI capabilities that differentiate an OpCo’s modernized applications: GenAI features, agentic workflows, and ML-powered functionality, integrated safely and reliably. You work side by side with the Technical Lead / Architect, contribute as a strong full-stack engineer, and help the whole pod raise its game in AI-native development.

If you want to turn the frontier of applied AI into shipped, dependable product features, this role is for you.

A TechGrove AI Engineering Pod is a self-contained, AI-native software delivery team that we embed inside one of our Operating Companies (OpCos) to build, modernize, and ship production software. Each pod pairs senior engineering talent with agentic AI tooling such as Claude Code, plus the accelerators of Banyan’s AI Application Modernization Factory, to deliver at a velocity and quality bar a traditional team cannot match.

A pod is typically four to eight people, and OpCos add more pods as their ambitions grow. You will work as part of a tight, high-trust team with real ownership of what you build. Pods are delivered from Banyan’s India-based TechGrove.

• Ship production AI/ML features: Design, build, and deploy GenAI integrations, agentic workflows, RAG and GraphRAG systems, and ML-powered functionality into the OpCo’s modernized applications.
• Orchestrate and integrate LLMs: Implement LLM orchestration, prompt and context engineering, guardrails, and evaluation to deliver reliable, safe, production-grade AI.
• Engineer cloud-native services: Build scalable, secure, resilient services that host and serve AI capabilities, as a strong hands-on contributor.
• Build agents and tooling: Develop AI agents and tooling, including building and connecting MCP servers and tools, and champion agentic development practices across the pod’s SDLC.
• Own data and retrieval: Design the data, retrieval, and vector or knowledge-graph infrastructure that powers GenAI features, with quality, performance, and security built in.
• Own technical design and quality: Lead design and implementation for AI features, write clean, well-tested code, and uphold engineering standards, security, and observability.
• Collaborate and mentor: Partner with the Architect and senior engineers, and coach the team on AI/ML engineering best practices.
• Apply pragmatic judgment: Know where AI adds real value versus risk, and navigate a fast-moving landscape with a clear head.

• Experience: 8+ years in software engineering, with significant hands-on experience implementing ML/AI features in production.
• Production experience on AWS or Azure (required): You have designed and built cloud-native applications on AWS or Azure, including containers, serverless, and Infrastructure-as-Code (Terraform). Experience limited to GCP alone does not meet this requirement, although GCP in addition to AWS or Azure is a plus.
• Applied AI/ML: Demonstrated experience implementing GenAI capabilities (RAG and GraphRAG, LLM orchestration, prompt and context engineering, guardrail design) and/or ML features in production systems.
• Agentic and GenAI frameworks: Hands-on experience with GenAI frameworks (LangChain, LlamaIndex), multi-agent frameworks (crewAI, AutoGen, or similar), and developing AI agents and MCP servers and tools.
• Software engineering depth: Strong full-stack ability with solid SQL/No-SQL and API design skills.
• Modern tech stacks: Hands-on production experience with at least two of: .NET Core, Python, TypeScript, and Java.
• AI-native tooling: Fluency with AI-assisted development tooling (Claude Code or similar) as a daily, core part of your workflow.
• Engineering principles: Solid grounding in DDD, design patterns, clean code, test-driven development, and application security.
• DevSecOps: Experience with CI/CD (GitHub Actions, GitLab CI) and cloud-native observability.
• Collaboration: Strong communication and an ownership mindset, comfortable with ambiguity and rapid change.
• Education: Bachelor’s degree in Computer Science or a related technical field, or equivalent practical experience.

• Deep LLM and vector database expertise: hybrid vector databases, knowledge graphs, and advanced prompt and guardrail design for enterprise AI safety.
• Experience fine-tuning or evaluating models, building eval harnesses, or implementing LLM observability.
• Experience building and operating multi-tenant SaaS at scale.
• Prior experience modernizing or re-platforming enterprise or B2B applications.

• First 90 days: You have shipped your first production AI feature, established reliable patterns for orchestration, retrieval, and evaluation, and the pod has a clear, safe approach to building with LLMs.
• First 6 months: AI capabilities are a dependable, differentiating part of the product, with measured quality and guardrails, and the team is meaningfully more capable in applied AI because of your mentorship.

• Permanent home, long-term thinking: Banyan acquires great software businesses and keeps them for good. We invest for the long term, which means you build software meant to last, not to flip.
• Real ownership: You own outcomes end to end inside a small, senior team. The OpCo owns the “what”; the pod owns the “how.”
• AI-native by default: You work every day with agentic tooling and the AI Application Modernization Factory, not as a side experiment but as the core of how we build.
• Variety with stability: Pods work across a portfolio of real products and domains, so you keep learning, with the security of a permanent employer behind you.
• Growth and excellence: A Banyan Director of Engineering Excellence supports every pod with coaching, standards, and continuous improvement, so your craft compounds over time.

We have written this role to be honest about what it takes, not to describe a mythical perfect candidate. If you meet most of what is listed and you are excited by the work, we want to hear from you. Strong engineers grow into the rest, and we will help you do exactly that.

We are building diverse, inclusive teams, and we welcome applicants of every background. What matters most is the quality of your engineering, your curiosity about AI-native development, and your appetite to own outcomes.