All engineering roles
Open engineering roles from public job boards. Postings that mention AI tools come first.
131 postings
Epic Games · Montreal,Quebec,Canada; Cary, North Carolina, United States
…Engine, including engine-side C++/Python tooling and editor scripting Design and extend the cloud-hosted agent harness, services, and database that mediate between LLMs and the engine Work across the full stack (engine tooling,…
Epic Games · Vancouver,British Columbia,Canada; Cary, North Carolina, United States
…Engine, including engine-side C++/Python tooling and editor scripting Design and extend the cloud-hosted agent harness, services, and database that mediate between LLMs and the engine Work across the full stack (engine tooling,…
Epic Games · BLANK,BLANK,Multiple Locations; Cary, North Carolina, United States
…Engine, including engine-side C++/Python tooling and editor scripting Design and extend the cloud-hosted agent harness, services, and database that mediate between LLMs and the engine Work across the full stack (engine tooling,…
Epic Games · Cary,North Carolina,United States; Cary, North Carolina, United States
…Engine, including engine-side C++/Python tooling and editor scripting Architect and extend the cloud-hosted agent harness, services, and database that mediate between LLMs and the engine Work across the full stack (engine tooling,…
Bankdata · Silkeborg, Denmark; Aarhus, Denmark
…Java samt API’er og integrationer GenAI og agentsystemer: LLM’er, GenAI frameworks, agentsystemer, skills og agent harnesses Data engineering: Datapipelines, datakvalitet og dataintegritet Cloud og DevOps: Cloudplatforme, gerne Azure,…
Snowflake · US-CA-Menlo Park; US-WA-Bellevue (United States)
StaffRemoteLLM observabilityAI / agent harnessAgentic workflowsEvals / LLM evaluationRAGGuardrails+1 more…models, or agent architectures. - Familiarity with eval and observability tooling (e.g., Braintrust, LangSmith, Arize, Weave, Promptfoo) or experience building custom eval harnesses. - Experience with failure-mode…
Mapbox · Mapbox US (United States)
Data / ML engineeringAI / agent harnessMulti-agent systemsEvals / LLM evaluationMCP (Model Context Protocol)…You can describe a dataset you built and the failure it caught. - Working knowledge of more than one agent harness, and opinions about where each of them is weak. - Experience diagnosing latency in a distributed…
Lightspark · Los Angeles HQ (USA)
Software (general)AI / agent harnessAgentic workflowsMCP (Model Context Protocol)Prompt / context engineering…across the company. This role extends it to Sales, Finance, Ops, Legal, Compliance, and Partnerships: the agent harness, tool integrations, governed source-of-truth data, and the execution layer those teams run their workflows…
Iterativehealth · Cambridge, Massachusetts, United States, New York, New York, United States; 675 Massachusetts Ave, Floor 8, Cambridge, MA 02139; 120 5th Ave, 10th Floor, New York, NY 10011; 620 E Southlake Blvd, Southlake, TX 76092
…startups (strongly preferred) Experience with LLM application frameworks (e.g. RAG, agent orchestration, eval harnesses) Familiarity with clinical trial operations, disease classification, or patient identification…
Mapbox · Mapbox Helsinki (Finland)
Data / ML engineeringFinlandRemoteAI / agent harnessMulti-agent systemsEvals / LLM evaluationMCP (Model Context Protocol)…hallucinated arguments, silent partial success, unbounded loops. - Working knowledge of more than one agent harness, and opinions about where each of them is weak. - Strong Python or TypeScript, and comfort reading code in…
Braze · New York City
Software (general)AI / agent harness…the team includes: Developing the framework used to build our marketing message editors Enabling our in-app AI harness to create, modify, and test messages in our Drag-and-Drop message editors Migrating the navigation and…
OpenAI · Singapore
…customer problems and produce measurable business outcomes. - Work hands-on in code to build prototypes, evaluation harnesses, reference implementations, integrations, and production accelerators. - Make sound technical decisions…
Mapbox · Mapbox UK (United Kingdom)
Data / ML engineeringUnited KingdomRemoteAI / agent harnessMulti-agent systemsEvals / LLM evaluationMCP (Model Context Protocol)…hallucinated arguments, silent partial success, unbounded loops. - Working knowledge of more than one agent harness, and opinions about where each of them is weak. - Strong Python or TypeScript, and comfort reading code in…
Mapbox · Mapbox Germany
Data / ML engineeringGermanyRemoteAI / agent harnessMulti-agent systemsEvals / LLM evaluationMCP (Model Context Protocol)…You can describe a dataset you built and the failure it caught. - Working knowledge of more than one agent harness, and opinions about where each of them is weak. - Experience diagnosing latency in a distributed…
Xm · Limassol; Nicosia; Athens; Thessaloniki; Fully Remote - in other eligible locations
QA / Test / SDETCyprusRemoteAmazon BedrockVertex AILLM observabilityAI / agent harnessAgentic workflowsMulti-agent systems+6 more…to take a leading technical role in AI quality and test automation. You will work primarily with our AWS Bedrock AgentCore-based CX Agent Suite, a multi-agent customer-support system covering intent routing, FAQs,…
Perplexity · San Francisco; New York City (United States)
…data and user feedback to guide decisions. - Practical experience in one or more relevant areas, such as agent harnesses, tool use, context engineering, model evaluation, browser automation, or long-running task execution. -…
Aircallioinc · New York Office; San Francisco Office; Seattle Office; New York Office; San Francisco Office; Seattle Office
…oversight. Bonus / Nice-to-have Experience building on or securing LLM and agent systems — Azure foundry, Bedrock/AgentCore, LangGraph, guardrails, policy-based authorization, prompt-injection defense, and agent…
Veeda AI · Toronto; Zürich; Seattle; California (Canada)
…tracking at scale, logging video, 3D, and trajectory artifacts rather than only scalars. - You have built evaluation harnesses for generative or embodied models, where quality is a distribution rather than a pass/fail. - Full stack…
Anthropic · San Francisco, CA | New York City, NY; San Francisco, California, United States
…error recovery — and reworking all of it with every model release — Managed Agents pairs an Anthropic-built agent harness with production infrastructure for sessions, environments, tools, memory, and permissions, exposed through a…
Callosum · London (United Kingdom)
…or retrieval over structured artefacts such as codebases - Deep familiarity with the internals of SGLang, vLLM, or comparable inference serving frameworks - scheduler design, memory management, and execution pipelines…
Devoteam · Koerich, Capellen, Luxembourg
Data / ML engineeringLuxembourgAmazon BedrockVertex AIAzure OpenAIvLLMLocal / open-weight LLMsMistral+13 more…LLM Deployment : define deployment strategies across managed cloud AI services (Azure OpenAI, AWS Bedrock, GCP Vertex AI) and open-weight models (Llama, Mistral) served on GPU infrastructure via vLLM or TGI; LLMOps,…
Airbnb · San Francisco, CA; United States
…develop the AI agents and agentic capabilities that automate trust decisions, and create the benchmarks and evaluation harnesses that keep decision quality high as those agents take on more autonomy. We work on problems before the answer…
Cognition · San Francisco; New York City (United States)
…with no established playbook. You'll create the conventions that others will follow. - Build the Agent and Agent Harness: Work directly on Devin and the core agent infrastructure that powers it. This means building the agent loop,…
OpenAI · New York City (United States)
…- Distill deployment learnings into reference architectures, interoperability and integration patterns, evaluation harnesses, security controls, and reusable technical primitives for healthcare and other regulated enterprise…
OpenAI · Seattle (United States)
…- Distill deployment learnings into reference architectures, interoperability and integration patterns, evaluation harnesses, security controls, and reusable technical primitives for healthcare and other regulated enterprise…
OpenAI · San Francisco (United States)
…- Distill deployment learnings into reference architectures, interoperability and integration patterns, evaluation harnesses, security controls, and reusable technical primitives for healthcare and other regulated enterprise…
Supabase · Remote, Global
DevOps / SRE / PlatformRemoteLLM observabilityAI / agent harnessMCP (Model Context Protocol)Prompt / context engineeringVector databasesAI security…an MCP server, an evaluation harness, or writing on agent reliability that other practitioners cite. - LLM observability and cost instrumentation. Tracing agent runs, attributing spend per run, and building the queries that turn…
Anthropic · San Francisco, CA | New York City, NY; San Francisco, California, United States
DevOps / SRE / PlatformAI / agent harness…programming, or inference optimization Experience building experiment tracking, caching layers, or evaluation harnesses for research teams Experience with probes, interpretability, or classifier development Interest in the misuse…
Neptunenorth · Madrid; Madrid, Spain
…experience leveraging LLMs in production, and an in depth understanding of prompt design, agentic workflows, eval harnesses, retrieval pipelines, etc. Exposure to LLM inference stack fundamentals such as KV Caching, Speculative…
Aquablu · Amsterdam, Netherlands
…collaborator. One day you're cutting a deploy pipeline from hours to minutes; the next you're building the eval harness so a builder can trust a feature behaves before it ever reaches a user; the day after you're putting…