Lead AI Engineer (Agentic Systems)
You've put AI agents into production, measured how often they were right, and fixed them when they weren't.
- Location
- Remote
- Experience
- 5+ years
- Compensation
- Market-competitive, USD-pegged
- Languages
- Python (primary), Node.js, TypeScript, React
- AI / agents
- LLMs, RAG, embeddings, tool calling, structured outputs, multi-agent orchestration, evals
- Frameworks
- LangGraph, LangChain, Semantic Kernel, AutoGen or equivalent
- Backend
- FastAPI, REST, async Python, microservices, task queues
- Data
- PostgreSQL, Redis, vector databases, document stores, data pipelines
- Infra
- AWS, Azure or GCP, Docker, Kubernetes, CI/CD, infrastructure-as-code
Since 2019 we've delivered 100+ projects. We run full engineering teams for US-based SaaS products and build AI-powered platforms end to end. Our clients are now moving from LLM features to agents that plan, call tools and run multi-step work on their own, and we need a senior engineer to lead that work from architecture to production.
We're hiring a lead AI engineer to architect and build agentic AI systems: agents that reason, plan, use tools, keep memory and carry out long, multi-step tasks. You'll own the technical design of LLM-powered applications and agent platforms from first concept to production, and write much of the Python yourself.
This is a technical leadership role. You'll set the standards for how agents are built, tested, evaluated and monitored, mentor the engineers working on them, and work with product managers, ML engineers and data scientists to turn business requirements into AI systems that hold up at scale. You'll also help shape the wider AI strategy and roadmap.
- Architect and build agentic AI systems capable of reasoning, planning, tool use, memory and multi-step task execution
- Lead the technical design of LLM-powered applications and agent platforms, from concept through production
- Build robust, maintainable Python services, frameworks and APIs that run AI agents at scale
- Design orchestration patterns: multi-agent systems, workflow-based agents, tool calling, RAG and human-in-the-loop workflows
- Integrate LLMs with internal systems, APIs, databases, enterprise applications and external tools
- Develop strategies for agent memory, context and state management, planning and long-running tasks
- Build evaluation frameworks that measure agent accuracy, reliability, latency, cost and safety
- Design observability, monitoring, tracing and debugging for agents in production
- Define best practices for prompt engineering, model selection, structured outputs, tool use and agent reliability
- Lead and mentor engineers, evaluate new models and frameworks, and set engineering standards for testing, security, privacy, scalability and responsible AI
- Languages
- Python (primary), Node.js, TypeScript, React
- AI / agents
- LLMs, RAG, embeddings, tool calling, structured outputs, multi-agent orchestration, evals
- Frameworks
- LangGraph, LangChain, Semantic Kernel, AutoGen or equivalent
- Backend
- FastAPI, REST, async Python, microservices, task queues
- Data
- PostgreSQL, Redis, vector databases, document stores, data pipelines
- Infra
- AWS, Azure or GCP, Docker, Kubernetes, CI/CD, infrastructure-as-code
Non-negotiable.
- 5+ years of software engineering, with significant experience designing and delivering production systems
- Expert-level Python
- Strong grasp of software architecture, distributed systems, APIs, databases, concurrency and cloud-native development
- Hands-on experience building applications with LLMs and generative AI, deployed and operated in production
- Deep understanding of agentic AI: tool/function calling, orchestration, planning and reasoning, RAG, memory and state, multi-agent architectures, human-in-the-loop, structured generation, and LLM evaluation and observability
- Experience with modern AI/ML APIs and frameworks
- Strong Git, CI/CD, automated testing and modern engineering practice
- Experience with AWS, Azure or GCP
- Excellent system design and technical communication, in clear written and spoken English
- A track record as a technical leader who influences engineering direction across teams
- Agent frameworks such as LangGraph, LangChain, Semantic Kernel or AutoGen
- Vector databases and retrieval, FastAPI or similar Python API frameworks
- Docker, Kubernetes and infrastructure-as-code
- Distributed task queues, event-driven architectures and workflow orchestration
- Model serving, fine-tuning, embeddings, reranking and model evaluation
- AI systems built under strict security, compliance, reliability or data-privacy requirements
- You've built AI observability or evaluation platforms
- Open-source AI/ML contributions, technical publications, patents or other public technical leadership
- USD-pegged salary
- Fully remote work
- Paid Claude and Codex subscriptions
- Technical ownership of production agent systems, and a real say in the AI roadmap
- Real US SaaS products on modern stacks
- Annual performance-based reviews
Show us how you work.
No cover letter, no easy-apply. This is what we read:
- Your CV
- Your GitHub or portfolio
- A short note on an AI agent system you designed and put into production
- A short video (2–3 minutes) walking us through it. Required: applications without one are rejected
Trouble with the form? Email hello@dataspecc.com.