Engineer II, AI Automation

Join our team and help shape the future of connectivity indoors and outdoors. Together, let’s push the boundaries of technology and advance sustainable networks worldwide.

How You’ll Help Us Connect the World:

As an AI Automation Engineer, you will design, build, and run the LLM-powered and agentic systems that are changing how we work. You will ship production AI services that automate triage, classification, and decision support for our operations and support teams, and build the AI-driven SDLC tooling that makes our engineers faster—from custom coding-assistant skills and agents to automated evaluation and guardrails.

This is a hands-on role where you are expected to pair strong backend engineering with practical GenAI skills. You will be based in Bangalore, Karnataka.

You Will Make an Impact By:

Designing, building, and deploying LLM-powered and agentic workflows that automate triage, classification, and decision support for operational and support teams.

Developing production-grade backend services in Python - REST APIs and async pipelines—with solid prompt engineering, retrieval-augmented generation (RAG), and evaluation practices.

Partnering with cross-functional stakeholders to turn ambiguous business problems into reliable, well-documented AI solutions.

Building and maintaining AI-driven SDLC automation such as spec-to-code generation, AI-assisted code review, and automated test generation embedded in everyday engineering workflows.

Developing and extending AI coding-assistant tooling for engineering teams, including Claude Code skills and plugins, custom agents, and reusable prompt templates.

Designing evaluation frameworks and guardrails for AI-generated output, using golden datasets, automated and LLM-as-judge scoring, and regression suites so quality is measured, not assumed.

Monitoring and improving model quality, reliability, latency, and cost through observability, guardrails, and feedback loops.

Maximizing token yield without compromising quality by delivering more useful, accepted output per token spent, with every optimization backed by evals that prove quality holds.

Optimizing multi-agent workflows for speed, cost, and token usage through context management, parallel execution, and prompt optimization.

Cutting token waste with tight context curation, prompt caching, right-sized model routing, and concise structured outputs.

Tracking token efficiency as a first-class metric including cost per successful task and tokens per accepted output while setting token budgets for agentic workflows.

Owning the delivery lifecycle in a DevSecOps environment (CI/CD, containerization, and automated testing from first commit to production)

Applying secure-by-default practices to every service you ship, including secrets management, access control, and dependency scanning.

Required Qualifications for Consideration:

Bachelor’s or Master’s degree in Computer Science, Information Science, Electronics Engineering, or a related field.

3 to 5 years of professional software engineering experience, including at least 1 to 2 years building LLM-powered applications that run in production.

Track record of taking AI features from prototype to production, with measurable impact on quality, speed, or cost.

Hands-on experience with LLM APIs (Anthropic Claude, OpenAI) including prompt engineering, tool/function calling, and structured outputs.

Practical experience building RAG systems: chunking, embeddings, vector databases (FAISS, Chroma, pgvector), hybrid search, and re-ranking.

Experience with agent frameworks such as LangGraph, the Claude Agent SDK.

Familiarity with LLM evaluation and guardrail techniques: golden datasets, LLM-as-judge, red-teaming, and mitigation of hallucination and prompt injection.

Daily use of AI coding assistants such as Claude Code, GitHub Copilot, or Cursor.

Strong Python skills, including asyncio, type hints, and testing with pytest.

Experience designing REST APIs with frameworks such as FastAPI or Flask.

Experience with async, event-driven pipelines and queues (e.g., Celery, Kafka, Redis), plus working knowledge of SQL and NoSQL data stores.

Experience with CI/CD (Jenkins, GitHub Actions, or GitLab CI), Docker, Kubernetes.

Working knowledge of secrets management (e.g., Vault, AWS Secrets Manager), role-based access control, and dependency and container scanning (e.g., Snyk, Trivy).

Experience with observability tooling (e.g., OpenTelemetry, Prometheus, Grafana, Langfuse) for tracing, metrics, and cost tracking of LLM applications.

Hands-on experience with a major cloud platform, preferably AWS.

You Will Excite Us If You:

Have experience building AI for IT operations or customer support

Are experienced in reducing LLM cost and latency at scale through prompt caching, model routing, batching, and context compression

Have applied AI tooling to large C/C++ codebases.

Have exposure to telecom or 4G/5G RAN, or a strong interest in learning the domain

Why ANDREW?

Explore exciting career opportunities at ANDREW, part of the Amphenol family. With a legacy of over 85 years in wireless innovation, we empower mobile operators and enterprises with cutting-edge solutions.

ANDREW, an Amphenol company, is proud of our reputation as an excellent employer. Our focus is to provide the highest level of support and responsiveness to both our employees and our customers, the world's largest technology companies. ANDREW offers the opportunity for career growth within a global organization. We believe that our company is unique in that every employee, regardless of his or her position, has the ability to positively impact the business.

ANDREW is an “Equal Opportunity Employer” - Minority/Female/Disabled/Veteran/Sexual Orientation/Gender Identity/National Origin

For additional company information please visit our website at https://www.andrew.com/