Developer

Job Summary

To develop and deliver codes for the work assigned in accordance with time| quality and cost standards.

HCLTech is looking for a hands-on AI Builder to develop the AWS DevOps Agent — an agentic Generative AI capability that automates root-cause analysis (RCA) and other DevOps operations tasks across AWS environments — under the direction of the Enterprise AI Architect. This role sits at the build layer of GenAI delivery: taking approved architecture and turning it into working, tested, production-ready agent capabilities (RCA reasoning chains, log/metric correlation, remediation recommendations, deployment risk scoring, incident summarization) using AWS-native AI and operations services integrated with LLM orchestration. You will work closely with the Enterprise AI Architect and DevOps/SRE teams to ship reliable, trustworthy automation quickly and to a consistent quality bar.

Key Responsibilities

• Build and implement the AWS DevOps Agent's core RCA capability: ingest logs, metrics, traces, and deployment events (Amazon CloudWatch, AWS X-Ray, AWS CloudTrail) and reason over them to identify likely root causes of incidents.
• Develop agentic workflows and tool-calling integrations (Amazon Bedrock Agents, LangChain/LangGraph, or equivalent) that let the agent query AWS services, run diagnostics, and propose or execute remediation actions.
• Extend the agent with additional DevOps capabilities beyond RCA: anomaly/drift detection, deployment risk scoring, change-impact analysis, incident summarization, and automated runbook execution.
• Integrate the agent with AWS-native operational tooling — Amazon DevOps Guru, AWS Systems Manager Automation, CloudWatch Alarms/Logs Insights, AWS Config, and CI/CD pipelines (CodePipeline/CodeBuild/CodeDeploy) — following architecture provided by the Enterprise AI Architect.
• Develop and tune prompts, function/tool definitions, and retrieval pipelines so the agent produces accurate, explainable RCA narratives and recommendations within latency and cost targets.
• Implement guardrails, human-in-the-loop approval gates, and safe-action boundaries so the agent can suggest or perform remediation without making unsafe autonomous changes to production.
• Write unit, integration, and evaluation tests for agent capabilities; track RCA accuracy, false-positive rate, mean-time-to-diagnosis, and other quality metrics.
• Support production readiness: contribute to observability of the agent itself (its own logs, traces, cost tracking), deployment scripts, and rollback procedures under architect guidance.
• Troubleshoot and resolve issues in existing agent capabilities, including reasoning regressions, tool-call failures, and integration issues with AWS services.
• Document agent capability build decisions, prompt/tool version history, and configuration details to keep the agent maintainable and auditable.
• Collaborate with the Enterprise AI Architect and DevOps/SRE stakeholders on feasibility checks, effort estimates, and technical trade-offs during agent capability design.
• Reuse and extend approved architecture patterns and reference implementations for agentic DevOps automation rather than building one-off solutions.

Skill Requirements

• 3-7 years in software/AI engineering, with hands-on experience building Generative AI or agentic applications, ideally in a DevOps, SRE, or observability context.
• Practical experience with agent frameworks and tool/function calling (Amazon Bedrock Agents, LangChain/LangGraph, Semantic Kernel, or equivalent) and prompt engineering for reasoning tasks such as RCA.
• Working knowledge of AWS observability and operations services: CloudWatch (Metrics, Logs Insights, Alarms), X-Ray, CloudTrail, AWS Config, and ideally Amazon DevOps Guru.
• Strong Python skills; comfort working with AWS SDK (boto3) integrations, REST APIs, and orchestration frameworks.
• Experience correlating logs, metrics, and traces to diagnose incidents (manually or through tooling), with an understanding of common failure modes in distributed/cloud systems.
• Understanding of basic security and safe-automation practices for agentic systems (action approval gates, least-privilege execution, audit logging) as defined by architecture guidelines.
• Familiarity with testing and evaluation approaches for GenAI and agent outputs (automated evals, human-in-the-loop review, regression testing on RCA accuracy).
• Working knowledge of CI/CD on AWS (CodePipeline/CodeBuild/CodeDeploy or equivalent) and basic infrastructure-as-code concepts.
• Ability to work from architecture and design specifications, ask clarifying questions, and flag feasibility issues early.
• Good written communication for documentation and handoffs to the architecture and DevOps/SRE teams.

Other Requirements

• Experience building or operating incident-management/RCA tooling (PagerDuty, Datadog, New Relic, Amazon DevOps Guru) prior to or alongside GenAI work.
• Exposure to multi-agent orchestration for complex diagnostic workflows (e.g., a triage agent handing off to a remediation agent).
• Familiarity with LLMOps/MLOps tooling (MLflow, Amazon SageMaker, Bedrock evaluation tools) for monitoring agent performance in production.
• Experience with GitHub Copilot or similar AI-assisted engineering tools in the SDLC.
• Basic understanding of Responsible AI concepts (bias, hallucination mitigation, explainability) as applied to automated RCA and remediation recommendations.