AI Engineer

Bangalore, India; Hyderbad, India·Posted today

llmtypescriptpythonnodenode.jsawsgcplangchainrag

<div class="content-intro">Recruitment Fraud Alert We’ve learned that scammers are impersonating Commvault team members—including HR and leadership—via email or text. These bad actors may conduct fake interviews and ask for personal information, such as your social security number.   What to know: <ul> <li>Commvault does not conduct interviews by email or text.</li> <li>We will never ask you to submit sensitive documents (including banking information, SSN, etc) before your first day.</li> </ul> If you suspect a recruiting scam, please contact us at <a href="mailto:wwrecruitingteam@commvault.com">wwrecruitingteam@commvault.com</a>    About Commvault  Commvault (NASDAQ: CVLT) is the gold standard in cyber resilience. The company empowers customers to uncover, take action, and rapidly recover from cyberattacks – keeping data safe and businesses resilient. The company’s unique AI-powered platform combines best-in-class data protection, exceptional data security, advanced data intelligence, and lightning-fast recovery across any workload or cloud at the lowest TCO. For over 25 years, more than 100,000 organizations and a vast partner ecosystem have relied on Commvault to reduce risks, improve governance, and do more with data. </div>AI & Automation Engineer - Marketing A software engineer first - who builds the automation pipelines and AI agents that run inside Marketing's workflows. You make AI reliable in the real world: prototype fast, harden for production, own it end to end. 01 THE ROLE You're a software engineer first - clean code, solid systems, reliability that holds in production. On top of that,  you bring AI into Marketing's day-to-day: automation pipelines that turn campaign, content, and lead data  into action, and AI agents that reason, use tools, and take action inside marketing workflows. You rapidly  prototype to prove what works, then harden it for production. You own the full stack: API design, cloud infra, AI integration, business rules, evaluation, monitoring, and the  feedback loops that make systems smarter. Marketing is the domain you serve - you don't need to come in  fluent in it, but you should be curious enough to learn how campaigns, content, and lead scoring actually  work. 02 WHAT YOU'LL OWN ▪ Re-engineer the process before automating it - map the marketing workflow, cut the steps that  shouldn't exist, then design the automation around the improved process, not the old one. ▪ Build end-to-end automation - connect marketing inputs (leads, campaign data, content requests,  forms) to outputs like scored leads, personalized content, and routed campaigns, through an AI  processing layer. ▪ Design and build AI agent systems - from single-agent task executors (e.g. drafting campaign copy,  triaging inbound leads) to multi-agent pipelines where agents specialise, coordinate, and hand off. ▪ Design human-in-the-loop governance and guardrails - confidence thresholds, approval gates,  escalation paths, and hard limits on what an agent may do unsupervised, including brand and  compliance-sensitive content. ▪ Run AI operations - monitor for drift and hallucination, manage prompt and model versions, control  token cost, and catch quality regressions before marketing users do. ▪ Integrate Azure OpenAI and other LLM services into production - prompt design, context management,  response validation, tool use, and function calling. ▪ Build RAG pipelines - ingestion, chunking strategy, embedding, vector store management, and retrieval  tuning over marketing content and knowledge assets. ▪ Build and run the REST APIs and Azure services behind every pipeline, integrate marketing and CRM  systems via their APIs and standard protocols like MCP, and write the business-rule and validation layers  on top of AI output. ▪ Tie every system to a business metric - pipeline contribution, campaign performance, or content velocity  - instrument it, and report the impact you actually delivered. 03 WHAT YOU BRING ▪ Strong fundamentals - algorithms, data structures, system design. Tested thoroughly in every round. ▪ Clean, maintainable, well-tested code a teammate can extend six months later. ▪ Shipped projects end-to-end - design, build, test, deploy, monitor, maintain. ▪ Comfort with ambiguity - incomplete requirements, surprising systems, figure-it-out energy. ▪ Clear communication of technical decisions to non-technical stakeholders - marketing will be your  primary partner. ▪ A real track record of self-teaching - our tools will be different in two years. ▪ Intellectual honesty about AI limits - you say clearly what AI can and can't reliably do, and design  accordingly. ▪ Patience with non-determinism - AI fails in ways hard to reproduce; you work it systematically instead of  shipping broken things. ▪ Marketing literacy - you know what a funnel, a campaign, an MQL/SQL, and a lead score are. You know  enough so that a marketing SME doesn't have to explain the domain from scratch.  04 THE STACK Core Engineering · Must have Any OOP language (Python, C# or Java), REST API design, async programming, Azure Functions / App Service  (or GCP / AWS equivalent), Service Bus, SQL & data modelling, Git · CI/CD, testing AI & Agents · Must have Azure OpenAI (or equivalent) / LLM APIs, prompt engineering, classification & extraction, output validation,  LangChain / LangGraph, Semantic Kernel, tool use / function calling, RAG implementation, MCP servers &  tools, agent evaluation AI Operations & Governance · Must have Human-in-the-loop design, agent guardrails & autonomy limits, drift & hallucination monitoring, prompt /  model version management, token cost control, eval frameworks & regression tests, outcome  instrumentation Bonus · Nice to have Exposure to marketing or CRM platforms (Salesforce, Dynamics 365, marketing automation tools), vector DB  internals, LLM observability, fine-tuning, TypeScript / Node.js - none required, all a plus. 05 HOW WE HIRE ▪ 01 · HR - Screen: Communication, culture fit, work ethic. ▪ 02 · HackerRank Test. 85% is the cutoff percentage. ▪ 03 · Tech - Build with us: System design + live coding across automation & agents. Video on, computer &  solid internet required. ▪ 04 · HR - Final round: Offer & wrap-up. LOCATION - Hyderabad / Bengaluru HOURS - 12:00–9:00 PM IST EXPERIENCE - 2–4 years TEAM - Marketing Technology Show us something you've shipped. A GitHub repo, a deployed system, or a detailed write-up of a real project beats any certification. If clean  code and hard problems are your thing - you belong here.<div class="content-conclusion"><div class="jd-info au-target" data-ph-at-id="jobdescription-text" data-ph-id="ph-page-element-page5-4e8TMw"> Commvault is an equal opportunity workplace and is an affirmative action employer. We are always committed to equal employment opportunity regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity or Veteran status and we will not discriminate against on the basis of such characteristics or any other status protected by the laws or regulations in the locations where we work. Commvault’s goal is to make interviewing inclusive and accessible to all candidates and employees. If you have a disability or special need that requires accommodation to participate in the interview process or apply for a position at Commvault, please email <a href="mailto:accommodations@commvault.com" target="_blank">accommodations@commvault.com</a> For any inquiries not related to an accommodation please reach out to <a href="mailto:wwrecruitingteam@commvault.com" target="_blank">wwrecruitingteam@commvault.com</a>.   <a href="https://www.commvault.com/privacy-policy">Commvault's Privacy Policy </a> </div>  </div>

Apply for this role