> Markdown version of [/jobs/ext/3958435-senior-software-engineer-it](https://www.wearedevelopers.com/jobs/ext/3958435-senior-software-engineer-it). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Senior Software Engineer - IT - **Company:** Dell Technologies Inc. - **Location:** United States - **Experience:** Expert - **Contract:** Internship / Graduate position - **Skills:** Application Programming Interfaces (APIs), Elasticsearch, Python (Programming Language), PostgreSQL, Reliability Engineering, Datadog, Data Logging, Large Language Models, Grafana, Langfuse — LLM Observability and Analytics Platform, Kubernetes, Splunk, Docker - **Published:** October 27, 2026 - **Apply:** https://www.jobmonkeyjobs.com/career/28138342/Senior-Software-Engineer-It-Any-Bangalore-1312 ## About the Role * 5+ years in site reliability engineering, production operations or platform engineering, with direct responsibility for systems in production. * Explicit, hands-on knowledge of performance-measuring and observability tools such as OpenTelemetry, Grafana, Datadog or Splunk - this is mandatory, not preferred, for this role. * Python (or equivalent) proficiency to build monitoring tooling, automate operational tasks and read the code of the systems being operated. * Working knowledge of Kubernetes and Docker for container operations, with infrastructure-as-code familiarity, plus PostgreSQL and Elasticsearch for metrics/log storage and querying. * Cost and consumption tracking for metered services, including LLM token accounting, usage attribution and forecasting against quota or budget. Desirable Requirements * Direct experience with LLM tracing tools such as LangFuse for agent-run performance measurement, in addition to general observability platforms. * Experience with LLM provider rate limits, quota management and multi-provider fallback, and runbook automation or failure-injection testing. ## Description * You will keep agentic security workflows operating reliably in production, restoring service quickly when performance degrades. * You will build the observability and performance-measurement layer that shows, in real time, how every workflow is running and what it costs. * You will make monitoring a routine step for every new workflow rather than a bespoke effort each time. You will: * Operate agentic security workflows in production, sustaining availability, completion rate, latency and cost within defined service levels, using performance-measuring tools such as OpenTelemetry, Grafana, Datadog or Splunk. * Build and operate the observability layer: execution traces across agent steps and tool calls, structured logging, health signals and dashboards, backed by PostgreSQL and Elasticsearch for metrics and log storage. * Measure token consumption and cost per agent, workflow and run using tools such as LangFuse, reporting actual cost against expectation so overruns are visible early. * Detect agent-specific failure modes - stalled or looping executions, tool call failures, credential and upstream API changes, silent quality degradation - and respond to restore service on Kubernetes/Docker-based deployments. * Triage production issues to root cause, escalate defects to the building team with traces and reproduction detail, and maintain runbooks so operational knowledge does not depend on one person.