> Markdown version of [/jobs/ext/2093938-ai-systems-specialist](https://www.wearedevelopers.com/jobs/ext/2093938-ai-systems-specialist). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# AI Systems Specialist

- **Company:** Perk
- **Location:** Barcelona, Spain
- **Contract:** Permanent contract
- **Skills:** Application Programming Interfaces (APIs), Artificial Intelligence, Application Frameworks, Automation of Tests, Software as a Service, Data Integrity, Fault Tolerance, Python (Programming Language), Salesforce.Com, Software Engineering, Systems Integration, Workflow Management Systems, AI Infrastructure, Large Language Models, Multi-Agent Systems, Zapier, Data Layers, Build Tools, Data Objects, Webhooks, Software Version Control, Data Pipelines, Automation Anywhere, Api Management
- **Published:** August 16, 2026
- **Apply:** https://es.trabajo.org/oferta-4021-55eaa4a16d8eb3978b4aef7cf04adc08

## About the Role

issues before they reach Sales. - Integrate with Salesforce at depth, including custom objects, flows, and automation, to ensure agent outputs are reflected accurately in the CRM and available to downstream reporting. Sales & GTM Team Enablement - Partner with GTM leadership to identify where manual effort can be replaced by intelligent automation, from account research and signal triage to follow-up sequencing and CRM hygiene. - Translate business problems into scoped, shippable agent systems, and surface realistic timelines and trade-offs rather than over-promising on what AI can reliably do today. - Build tooling and interfaces that give non-technical teams visibility into agent activity and confidence in the outputs they are acting on. - Rapidly diagnose and resolve failures in production systems to minimise disruption to selling time. Technical Standards & Team Leverage - Set and maintain engineering standards for how AI systems are built within Revenue
Systems: version control practices, documentation, testing patterns, and deployment hygiene. - Share frameworks, code, and learnings with the broader GTM Engineering team to drive consistency across the function. - Provide clear feedback to inform roadmap prioritisation, surfacing where automation is working, where it is brittle, and where human judgment still beats a model. What We're Looking For Must-Have - 5+ years in a technical role building production systems, whether GTM Engineering, Revenue Systems, or software engineering with strong GTM exposure in a B2B SaaS environment. - Hands-on experience building multi-step AI agents using Python and LLM APIs (Claude, OpenAI, or equivalent) in a production context-beyond prompt experimentation: deployed systems with real dependencies and users. - Strong Python engineering fundamentals. You can build clean, maintainable, testable code and you care about the difference between a working prototype and a
reliable system. - Direct experience integrating APIs, webhooks, and event-driven flows using automation tooling (Make, n8n, Zapier, or equivalent) alongside custom Python solutions. - Salesforce proficiency at the configuration layer: custom objects, flows, reporting, and data integrity logic. - A track record of translating ambiguous business problems into scoped, delivered technical solutions. You know how to size a problem, push back when scope is unrealistic, and ship. - Strong instincts for system design. You build things that are observable, recoverable, and do not create maintenance debt for the people who come after you. Strong Advantage - Experience designing agent frameworks or shared infrastructure patterns that other developers build on top of. - Familiarity with the GTM tooling landscape, including Clay, Outreach, Gong, and Apollo, and how to connect agent outputs into that stack effectively. - Exposure to conversation intelligence platforms (Gong

## Description

We value curiosity, purpose, and mindset, not just knowledge, to unlock the power in your potential. Our talent team brings together leading minds from the travel and SaaS industries, representing over 70 countries. If you're excited about having a real impact and shaping how millions of people experience work, we'd love you on the team. Visit www.perk.com to learn more. About the Role At Perk, we're building AI infrastructure that changes how a revenue team operates. Not a layer of tooling bolted on top of existing workflows, but a foundation that makes intelligent automation the default way work gets done. The Revenue Systems team sits at the centre of this, responsible for the systems, data, and agents that power how Sales and GTM operate. The AI Systems Specialist is a senior individual contributor and the technical authority on agentic AI within Revenue Systems. This is a role for someone who has moved past experimenting with LLMs and is building production-grade, multi-step
agent systems that replace real manual work. Where the GTM Engineering function owns Clay-based outbound and enrichment infrastructure, the AI Systems Specialist owns the agent layer: the Python-backed workflows, API integrations, and reusable frameworks that let the team ship AI automation faster and more reliably over time. The 12-month goal is not a list of shipped workflows. It is a reusable agent framework that the broader Revenue Systems and GTM Engineering team builds on top of. That means strong engineering instincts alongside strong business judgment: knowing which problems are worth automating, how to build systems that do not break quietly, and how to make your work accessible to non-engineers who depend on it. The role reports to the Senior Manager, Revenue Systems, and works closely with the GTM Engineering team and Sales leadership. It is a highly visible role with a direct line of sight to commercial outcomes. AI is embedded in how we work across every
function. The tools you will use day-to-day are AI-enabled so you can move faster and spend more time on the work that matters. For this role specifically, working with AI is not a part of the job-it is the job. We will expect you to have strong opinions on where LLMs are actually useful today versus where they are unreliable, how to build systems that behave predictably in production, and where human judgment still beats automation. We will ask about how you use AI and how you build with it in your interviews, not to test tool knowledge, but to understand how you think. What You'll Do Agentic Workflow Design & Build - Design, build, and maintain multi-step AI agents using Python and LLM APIs (Claude, OpenAI) that automate research, qualification, routing, and decision-making across the revenue stack. - Architect agent systems that are robust in production: with clear error handling, observability, and graceful degradation when upstream data or APIs misbehave.

 - Build the reusable agent framework, including shared abstractions, tooling patterns, and integration modules, that allows the wider Revenue Systems and GTM Engineering team to ship AI workflows without starting from scratch each time. - Evaluate and integrate emerging agentic tooling and patterns, bringing well-reasoned recommendations on when to build vs. buy and which abstractions are worth investing in. Data Pipelines & Systems Integration - Build and maintain data pipelines that connect Salesforce, outbound tools, enrichment providers, and internal systems, ensuring the clean, timely data that agents depend on is always available. - Engineer API integrations, webhook-driven triggers, and event-based flows using Python, Make, and equivalent automation tooling without requiring engineering support for each new connection. - Own the data layer that underpins AI decision-making: designing schemas, maintaining data quality, and building monitoring logic that catches


## Related Videos

- [HR ROBO SAPIENS: Decoding AI Agents and Workflow Automation for Modern Recruitment](https://www.wearedevelopers.com/videos/1470-hr-robo-sapiens-decoding-ai-agents-and-workflow-automation-for-modern-recruitment)
- [Agentic employees in world's most downloaded FinTech app](https://www.wearedevelopers.com/videos/100123-agentic-employees-in-world-s-most-downloaded-fintech-app)
- [Modern Data Architectures need Software Engineering](https://www.wearedevelopers.com/videos/1030-modern-data-architectures-need-software-engineering)
- [Beyond Webhooks: The Future of Scalable API Event Delivery](https://www.wearedevelopers.com/videos/100312-beyond-webhooks-the-future-of-scalable-api-event-delivery)
- [Building a framework-independent component library](https://www.wearedevelopers.com/videos/1679-building-a-framework-independent-component-library)
- [The Vonage Trivia Voyage: Quiz Your Way to the Top!](https://www.wearedevelopers.com/videos/761-the-vonage-trivia-voyage-quiz-your-way-to-the-top)

## Related Articles

- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)
- [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer)
- [From Prototype to Production: Build AI Agents with This Free 4-Course Learning Path](https://www.wearedevelopers.com/magazine/655-from-prototype-to-production-build-ai-agents-with-this-free-4-course-learning-path)
- [Navigating the AI Shift](https://www.wearedevelopers.com/magazine/629-navigating-the-ai-shift)
- [Coffee with Developers - Maria Apazoglou - Making AI understandable for all in production](https://www.wearedevelopers.com/magazine/475-coffee-with-developers-maria-apazoglou-making-ai-understandable-for-all-in-production)
- [Graph and AI Trends 2026: Why Is AI Running but Not Yet Delivering?](https://www.wearedevelopers.com/magazine/680-graph-and-ai-trends-2026-why-is-ai-running-but-not-yet-delivering)