About dave.io
dave.io is an AI-native cloud infrastructure service that gets infrastructure work done end to end.
We believe the next evolution of cloud infrastructure is moving from AI tools that help engineers do the work to AI-native services that take ownership of it. Dave plans, executes, and manages infrastructure work end to end, delivering outcomes, not just assistance.
But real infrastructure demands real accountability. Dave combines AI-native automation with a team of expert cloud engineers who stay in the loop, reviewing and approving every step and bringing the oversight, context, and human judgment needed to operate infrastructure with confidence.
The result: teams can dramatically increase their infrastructure output without scaling their headcount.
We’re a well-funded, seed-stage startup, and dozens of organizations already run Dave in production. Our team comes from monday.com and JFrog, and we’re building alongside the broader cloud ecosystem, with partners including AWS, GCP, Azure, compliance platforms, and leading MSPs.
We’re moving fast, and we’re just getting started.
The role
As an AI Engineer at Dave, you’ll build the intelligence at the core of our product: the systems that enable Dave to understand infrastructure problems, plan how to solve them, use tools, take action, and learn from the outcome.
You’ll work on the hard problems behind making AI agents actually useful in real production environments: reasoning across complex infrastructure context, planning multi-step tasks, choosing and using the right tools, evaluating outcomes, recovering from failures, and knowing when human judgment is required.
This is not a research-only role and it’s not about building simple LLM wrappers. You’ll own AI capabilities from experimentation through production, working closely with Product, Engineering, and our cloud experts to turn real infrastructure workflows into reliable agentic systems.
Because Dave operates on real customer infrastructure, reliability matters as much as capability. You’ll help us build AI systems that are not only powerful, but also observable, measurable, controllable, and safe enough to take on meaningful infrastructure work.
Responsibilities
- Design, build, and improve the AI agents and workflows that power Dave.
- Build systems that enable agents to understand infrastructure context, plan complex tasks, use tools, execute multi-step workflows, and evaluate their results.
- Develop and improve agent capabilities across reasoning, planning, tool use, context management, retrieval, memory, and orchestration.
- Build reliable integrations between AI agents and infrastructure tools, APIs, cloud platforms, repositories, CI/CD systems, observability platforms, and other parts of the modern DevOps stack.
- Design evaluation frameworks and benchmarks that measure agent quality, task completion, reliability, regressions, and failure modes.
- Analyze real-world agent failures and systematically improve prompts, context, tools, workflows, models, and system architecture.
- Build feedback loops that turn production usage and expert review into measurable improvements in agent performance.
- Design guardrails, validation mechanisms, approval flows, and human-in-the-loop systems that allow Dave to operate safely in customer environments.
- Build observability and debugging capabilities that help us understand what agents are doing, why they fail, and how their performance changes over time.
- Experiment with different models, architectures, prompting strategies, and agent patterns, and make pragmatic decisions based on quality, latency, reliability, and cost.
- Work closely with our cloud engineers to capture expert DevOps knowledge and translate it into capabilities Dave can execute autonomously.
- Partner with Product and Engineering to take new AI capabilities from early experimentation to production.
- Stay close to advances in LLMs and agentic AI, quickly evaluating new approaches and applying the ones that meaningfully improve the product.
Skills and experience
- 4+ years of software engineering, AI engineering, machine learning engineering, or similar experience, with strong production engineering fundamentals.
- Hands-on experience building production systems with LLMs and/or AI agents.
- Strong software engineering skills, preferably with Python, and experience building production-grade backend systems.
- Experience with agentic systems, including areas such as tool/function calling, multi-step workflows, orchestration, context management, retrieval, memory, or planning.
- Experience designing and running evaluations for LLM or agent-based systems and using them to drive measurable improvements.
- Strong understanding of the challenges of production AI systems, including nondeterminism, hallucinations, reliability, latency, cost, observability, and failure recovery.
- Experience integrating AI systems with APIs, tools, external systems, and structured and unstructured data.
- Ability to take an ambiguous problem, experiment quickly, measure results, and turn a promising approach into a robust production system.
- Strong debugging and analytical skills, especially when working with complex or nondeterministic systems.
- Product mindset: you care about whether the system actually solves the problem, not just whether the model produces an impressive output.
- High level of ownership and comfort taking ideas from first principles through experimentation, implementation, deployment, and iteration.
- Comfortable working in a fast-moving early-stage startup where the technology, product, and best practices are evolving quickly.
- Deep curiosity about AI agents and excitement about redefining how infrastructure work gets done.
Great to have
- Experience building agents that take real-world actions rather than only generating content or answering questions.
- Experience with LLM evaluation, tracing, observability, or experimentation platforms.
- Experience with RAG, embeddings, vector search, or other retrieval and context-management techniques.
- Experience with agent frameworks or orchestration systems such as LangGraph, LlamaIndex, or similar technologies.
- Familiarity with MCP and emerging standards for connecting AI agents to tools and external systems.
- Experience with cloud infrastructure, DevOps, developer tools, or infrastructure automation.
- Familiarity with AWS, GCP, Azure, Kubernetes, Terraform/IaC, CI/CD, or observability platforms.
- Experience designing human-in-the-loop systems, approval mechanisms, or safety controls for autonomous agents.
- Experience working at an early-stage or fast-growing startup.
Apply for AI Engineer
It takes a minute. Your LinkedIn profile is all we need to start.