Sr. Prompt Engineer

Job highlights

• Demonstrated experience building and improving production-oriented LLM applications,AI agents,RAG systems,semantic search,structured-output pipelines,or AI workflow automations. . Strong prompt-engineering judgment: able to design prompt and output contracts,diagnose failures,run controlled experiments,and choose when prompting alone is insufficient.

Job Summary

Business Technology, Pune, India/ Bangalore, India

Sr. Prompt Engineer

Responsibilities

This is a hands-on individual-contributor role for a prompt engineer who helps deliver dependable enterprise AI solutions for defined go-to-market and business problems. You will build on the AI Fabric - the shared foundation for governed execution, trusted context, reusable integrations, and observability - and use automation plus code to deliver workflows people can rely on.

You will work across prompts, models, context, data, tool and API integrations, workflow automation, evaluations, and operational controls. This is not a prompt-only role: you will apply established platform patterns to build and improve well-scoped solutions, partnering with stakeholders and the appropriate technical owners from use-case definition through launch, measurement, and continuous improvement.

Here's a closer look at this key role:

• Partner with stakeholders to clarify defined go-to-market and business use cases, then build and deliver enterprise AI solutions on the shared AI Fabric using established patterns, approved connections, and agreed scope. Contribute improvements rather than creating one-off or parallel implementations.
• Apply and iteratively improve established prompt, system-instruction, structured-output, tool-use, model-routing, and context-window practices so outputs are useful, grounded, predictable, and fit for a business workflow.
• Build and support n8n or comparable automation workflows using established patterns for triggers, APIs, webhooks, human approvals, retries, fallbacks, scheduled jobs, and error handling. Contribute Python, Strands SDK, or other lightweight code when a workflow needs approved custom logic or durable runtime behavior.
• Integrate AI solutions with enterprise systems and governed data sources such as Slack, Salesforce, Gong, Jira, Confluence, Google Workspace, Snowflake, AWS services, and internal APIs. Pull, join, and validate data from the appropriate tables while applying established access, data-handling, and audit standards.
• Apply established context and data standards to select the approved sources for a use case; shape, prioritize, and refresh context; manage context-window limits; support permission-aware retrieval; and keep response behavior grounded in source data. Escalate new source, access, or architecture decisions to the appropriate owners.
• Apply established decision criteria for LLM-driven and deterministic components. For example, use models for language understanding and generation while relying on SQL, rules, validated APIs, and calculation services for financial, policy, or data-accuracy-critical operations; escalate new architectural tradeoffs when needed.
• Apply existing evaluation and quality controls, including representative test sets, regression tests, LLM-as-judge or deterministic scoring, failure taxonomies, prompt and retrieval experiments, and release criteria. Contribute evidence and improvements to the shared evaluation approach.
• Use approved observability and operational practices for production solutions, including structured logs, outcome and quality metrics, token and cost visibility, audit trails, alerting, troubleshooting guidance, and feedback loops.
• Apply privacy, security, and reliability controls such as data minimization, PII redaction or placeholder restoration where appropriate, secrets management, schema validation, human-in-the-loop gates, graceful degradation, and least-privilege integration patterns.
• Work directly with senior go-to-market and business stakeholders, product, data, security, and engineering teams to clarify use cases, translate approved requirements into practical solution designs, surface risks and tradeoffs to the appropriate owners, and use observed usage to improve solutions.
• Contribute reusable technical modules, templates, and documentation that help future AI Fabric and n8n solutions be faster, safer, and easier to support.

What You'll Need

• 4+ years of experience in prompt engineering, software engineering, data or automation engineering, AI/ML engineering, solutions engineering, or a comparable hands-on technical role.
• Demonstrated experience building and improving production-oriented LLM applications, AI agents, RAG systems, semantic search, structured-output pipelines, or AI workflow automations.
• Strong prompt-engineering judgment: able to design prompt and output contracts, diagnose failures, run controlled experiments, and choose when prompting alone is insufficient.
• Working knowledge of workflow automation concepts and tools such as n8n or comparable platforms, including integrations, webhooks, error handling, retries, and operational support; able to learn and apply the team s n8n standards.
• Working proficiency in SQL and practical ability to write queries that select, join, aggregate, and validate data from tables. Working knowledge of relational databases, schemas, data quality, and the tradeoffs of using governed data sources.
• Ability to use Python and/or JavaScript or TypeScript for API integration, data transformation, automation, and AI workflow implementation.
• Experience integrating APIs, SaaS platforms, webhooks, cloud services, and enterprise data systems with appropriate authentication, secrets, logging, error-handling, and data-access practices.
• Experience applying context-engineering and retrieval concepts including established source-selection practices, embeddings, chunking, metadata, permission-aware retrieval, source attribution, refresh patterns, context-window management, and grounding techniques.
• A rigorous testing and evaluation mindset, including test datasets, regression testing, output validation, and measurable quality metrics.
• Ability to work directly and credibly with senior stakeholders: clarify a business use case, communicate solution design, implementation status, risk, and tradeoffs, and support adoption with both technical and non-technical partners.

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.