About Headroom

We build production AI for firms that need it to work.

Headroom Consulting builds production-grade AI systems for professional services firms doing $5M-$20M in revenue. We don't hand over a slide deck and walk away - we deploy fully operational AI workflows that save our clients 350-900 hours per month, train their teams to own the system, and measure every dollar of ROI.

We're a small, flat team that ships fast and holds itself to an uncomfortably high standard. If you join, you'll work directly with the founder and 2-3 other engineers who are just as obsessive about quality as you are.

The role

Architect, build, and deploy AI workflows - end to end.

You'll architect, build, and deploy AI-powered automation systems for our clients. That means scoping workflows, writing production code, integrating with the tools our clients already use (email, CRMs, project management, document systems), and making sure everything runs reliably after you hand it off.

This is not a research role. This is not a "build a chatbot and call it a day" role. You'll be building systems that replace 100+ hours of manual work per week per client - and you'll be accountable for that result.

What you'll do

• Design and deploy AI workflow automations using Claude Code as your primary development environment, alongside tools like n8n, Make, and custom integrations.
• Build custom AI/ML pipelines - prompt engineering, fine-tuning, API orchestration, retrieval-augmented generation, and multi-agent architectures.
• Write clean, production-grade Python and JavaScript that other engineers can maintain and extend.
• Scope and map client workflows to identify the highest-ROI automation targets.
• Deploy into client environments and ensure stability, monitoring, and graceful failure handling.
• Occasionally present technical solutions to clients and translate complex AI concepts into clear business language.
• Continuously improve internal tooling, templates, and deployment playbooks.

Qualifications

Non-negotiable. You will not be considered without these.

• 2+ years of hands-on experience building and deploying AI/ML systems in production - not just prototypes, not just notebooks, not just fine-tuning tutorials. Production. With users. Generating business outcomes.
• Deep fluency with large language models - advanced prompt engineering, knowing when to use RAG vs. fine-tuning vs. agents, and articulating the tradeoffs in your sleep.
• Proficiency with Claude Code - or demonstrated ability to work at the same speed and depth using AI-assisted development tools in a professional setting.
• Strong software engineering fundamentals - clean architecture, version control discipline, testing, CI/CD, error handling. If your code can't survive a production environment without you babysitting it, this isn't the right fit.
• Experience integrating AI into real business workflows - CRMs, email systems, document pipelines, project management tools. You've connected the dots between "AI capability" and "business process" and made the result work reliably.
• Automation platform experience - n8n, Make, Zapier, or equivalent. You know when to use a no-code tool and when to write a custom solution.
• Excellent written and verbal communication - you'll occasionally interface with clients, and you need to explain what you're building and why without hiding behind jargon.

Strongly preferred - will put you at the top of the list

• Experience building multi-agent systems or complex agentic workflows.
• Background in consulting, professional services, or agency environments where client delivery timelines are non-negotiable.
• Track record of measurably reducing manual work hours for a business (we want numbers, not vague claims).
• Experience with vector databases, embeddings, and semantic search at scale.
• Familiarity with workflow mapping or business process analysis.
• Contributions to open-source AI/ML projects, or published technical writing that demonstrates depth of thinking.

How to apply

Submit the following.

1. Your résumé or LinkedIn profile. The basics. We won't hire on this alone, but we want it.
2. A portfolio or 2-3 case studies. For each AI system you've built and deployed, show: the problem, your approach, the technical architecture, and the measurable result. "I built a chatbot" won't cut it. "I automated a 12-step client onboarding workflow that reduced processing time from 3 days to 20 minutes" will.
3. A short written response - under 500 words. Pick one workflow inside a professional services firm (accounting, legal, marketing, consulting - your choice) and describe how you'd automate it using AI. Be specific about the tools, the architecture, and the expected ROI.

Candidates who pass the portfolio review will be given a paid technical assessment - a real-world automation challenge that mirrors the work you'd do on the job.

Why this role is different

Real systems. High ownership. Direct access.

• You're building real systems, not demos. Every workflow you deploy saves a client hundreds of hours and hundreds of thousands of dollars.
• Small team, high ownership. No layers of management. No waiting for approval. You own the outcome from scoping to deployment.
• Direct access to leadership. You work directly with the founder - not through a project manager three levels removed.
• Skill compounding. You'll become an expert at the intersection of AI and business operations - one of the most valuable skillsets in tech right now.

Headroom Consulting is an equal opportunity employer. We evaluate candidates based on skill, proof of work, and alignment with our standards - nothing else.