ML & Agentic Systems Engineer [IC4]

Sourcegraph
Anywhere
Workplace: RemoteFull timeUSD 88,000 - 176,000 annuallyFunction: IT Operations (Systems/Network Admin)Skills: ["Technical leadership","Mentorship","Evaluation judgment","Autonomy","Customer-driven product thinking"]

Build and harden production agentic systems for code understanding, owning multi-step tool-using agent loops and the model/agent pipelines behind key product surfaces. Lead pragmatic evaluation strategy for agentic changes, deciding how and when to use evals, smoke tests, and metrics. Own model selection, upgrading, fine-tuning, and retrieval/context engineering to improve answer quality while controlling cost and latency—at staff scope on a fast, customer-driven team.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Sourcegraph
Sourcegraph
23 hours ago

ML & Agentic Systems Engineer [IC4]

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 1 hour agoStatus: Live

Job Summary

Build and harden production agentic systems for code understanding, owning multi-step tool-using agent loops and the model/agent pipelines behind key product surfaces. Lead pragmatic evaluation strategy for agentic changes, deciding how and when to use evals, smoke tests, and metrics. Own model selection, upgrading, fine-tuning, and retrieval/context engineering to improve answer quality while controlling cost and latency—at staff scope on a fast, customer-driven team.
Location: Anywhere
Workplace: Remote
Employment Type: Full time
Job Function: IT Operations (Systems/Network Admin)
Seniority: Mid level

Key Responsibilities

  • •Design and harden multi-step, tool-using agent loops and translate research into reliable, observable products at enterprise scale.
  • •Create evaluation strategy for agentic products, including when to use smoke tests/metrics and how to avoid noisy rigor.
  • •Own model selection, upgrades, and training/fine-tuning decisions, aligned to quality, cost, and latency constraints.
  • •Lead retrieval and context engineering (grounding models in customer code) including retrieval, ranking, context windows, and citations.
  • •Treat cost and latency as product features by profiling, distilling, caching, and right-sizing models to ship sustainably.

Pay and Benefits

Salary: USD 88,000 - 176,000 annually

Key Requirements

  • •Staff-scope skills across production machine learning, evaluation, and agent systems.
  • •Personally owned a production model lifecycle end-to-end, including training/fine-tuning and evaluation to rollout and monitoring.
  • •Designed multi-step agentic systems that are reliable, observable, and cost-bounded.
  • •Built strong evaluation judgment: representative datasets, baselines, error analysis, and release criteria connecting offline metrics to production.
  • •Strong software engineering ability to ship production services across an async-first, multi-service remote environment.
Skills:Technical leadershipMentorshipEvaluation judgmentAutonomyCustomer-driven product thinking
Tech Stack:GoTypeScriptGraphQLPostgresDockerLLMRetrievalRankingEmbeddingsFine-tuningDistillationPromptingTool-using agentsMulti-step agent loopsEvalsDashboardsGuardrailsCaching

Company Brief

Sourcegraph
Provides universal code search and intelligence for developers and engineering teams, enabling fast code navigation, cross-repository search, code intelligence, and developer productivity tooling for both cloud and self-hosted environments.
Industry: Developer Tools
Company Size: Large (251 to 1,000 employees)
Growth: Scaleup
Funding: Series D
Headquarters: San Francisco, United States
Founded: 2013
WebsiteLinkedIn