Senior AI Engineer (Low Code)

Ready to shape the future of data? Matillion is the intelligent data integration platform. We're changing how the world works with data – and we need driven, curious people who think big and move fast.  We built the Data Productivity Cloud to supercharge data productivity, and now we’re shaping the future of data engineering with Maia – our AI-powered virtual data engineers that help teams design, build, and manage data pipelines at unmatched speed. Join #TeamGreen, where the mission comes first, collaboration drives us forward, and everyone pulls in the same direction to make a dent in the universe bigger than ourselves. We are now looking for an AI Engineer to join our IT and Security team, based from our Hyderabad, India offie. Role Purpose Matillion is building an agentic workforce - AI-powered systems that autonomously tackle business problems across sourcing, prospecting, reporting, and operational copilots. This role exists to accelerate our AI adoption by designing, deploying, and iterating production-grade autonomous agents using low-code orchestration platforms. You'll sit at the intersection of business process and AI capability, translating messy real-world workflows into reliable, self-improving agent systems. This isn't about writing code from scratch—it's about wiring together LLMs, APIs, and data sources to solve problems at speed. Success means shipping agents that work, monitoring what breaks, and fixing it before anyone notices. What you will be doing

Agent Design & Orchestration

Build multi-step AI agents in platforms like Relevance.ai, Make.com or Langflow that autonomously perform research, data extraction, and decision-making

Design agent workflows that anticipate failure modes and include error handling, retry logic, and human escalation paths

Connect AI "brains" to business tools (CRM, Slack, email, databases) via APIs, webhooks, and MCP protocols

Develop, version-control, and iterate complex system prompts to ensure output consistency across production workloads

Implement testing frameworks to catch prompt drift and hallucination patterns before they reach production

Document prompt logic and decision trees so non-technical stakeholders can understand agent reasoning

Build Retrieval-Augmented Generation (RAG) pipelines that connect agents to internal knowledge bases (docs, wikis, databases)

Optimize retrieval logic to surface the right context without overwhelming token limits or introducing noise

Manage vector databases and ensure data freshness across agent memory systems

Partner with business process owners (Sales, Marketing, Ops) to translate workflows from plain-language descriptions into executable agent logic

Identify where automation adds value vs. where humans should stay in the loop

Challenge requirements that sound good but won't work in production

Analyze agent logs to identify failure patterns, logic gaps, and edge cases

Run A/B tests on workflow variations to measure success rate improvements

Iterate on "wiring" (workflow structure) and prompts based on real-world performance data

Work autonomously on small-to-medium complexity agents; support peers on larger initiatives

Document agent architectures, prompt libraries, and integration patterns for team reuse

Share learnings across the AI Engineering team to accelerate collective capability

What we are looking for

Technical / Role-Specific Capabilities

3-5 years building low-code automations and applications (e.g., Zapier, Make.com, n8n, Power Automate, or similar)

Hands-on experience with LLM orchestration platforms (Relevance.ai, Langflow, LangChain, or equivalent)

Demonstrated ability to design and deploy API integrations (REST, GraphQL, webhooks)

Working knowledge of LLM behavior, including prompt engineering techniques, token management, and hallucination mitigation

Experience building Retrieval-Augmented Generation (RAG) workflows or vector database integrations (Pinecone, Weaviate, ChromaDB, etc.)

Systems thinking: Ability to map business processes, identify bottlenecks, and design automated workflows that account for edge cases

Trial-test-iterate mindset: Comfortable shipping MVPs, monitoring performance, and iterating based on data

Experience with Model Context Protocol (MCP) or similar agent-to-tool communication frameworks

Familiarity with Python or JavaScript for custom function/API development

Previous work in data engineering, analytics, or business intelligence roles

Understanding of GDPR and data handling compliance in AI systems

Collaborative: Works effectively in teams and autonomously; shares knowledge openly and supports peers

Communicates Courageously: Challenges bad requirements tactfully; explains technical trade-offs to non-technical stakeholders in plain language

Passionate: Genuinely excited by AI's potential to solve problems (but realistic about its limitations)

Honorable & Caring: Takes responsibility for agent failures; considers ethical implications of automation on teams

Authentic: Admits when they don't know something; asks for help when stuck

Outcome-focused: Measures success by business results, not lines of code or number of agents shipped

Key Competencies

Execution - Delivers complete solutions with minimal supervision; anticipates common failure modes

Technical Depth - Deep expertise in low-code platforms; working knowledge of LLMs and APIs

Collaboration - Partners effectively with business teams to translate needs into agent workflows         

Problem-Solving - Debugs agent failures using logs and testing; iterates solutions based on data

Communication - Explains technical decisions clearly to non-technical audiences; documents work for reuse

Ownership - Takes end-to-end responsibility for agent reliability and performance