AI Solution Engineer
Confirmed on the employer's own hiring board on Oct 18, 2026. First seen by Alion on Mar 31, 2026. Keyloop scores B on the Alion truth index.
The Opportunity
Customer Enablement runs on data and on tooling. Our support, retention and operations teams need reporting, dashboards and automation to do their jobs - and because modern AI tooling makes building things fast, a good deal of it has been built locally, by the people who needed it, in an afternoon.
Much of it is genuinely useful. The problem is that it is not built to last: data is loaded by hand, nothing is scheduled, failures are silent, there is no documentation, and when the person who built it is on leave, it stops working. Your job is to take that work on and make it real engineering - automated, monitored, documented, properly hosted, and owned by the function rather than by an individual. Then to build what comes next that way from the start.
This is a hands-on software engineering role, and it is a broad one. You will work across the systems CE depends on - CRM, collaboration tools, data warehousing and reporting - building the integrations that remove manual steps, and using modern AI tooling such as Claude and Microsoft Copilot as part of how you build. You will own what you deliver end to end, from requirement through to running it in production, and you will work directly with the people who use it.
To be clear about the kind of AI work this is: this is applied engineering, not machine learning research or model development. You will be building software that uses AI tooling and LLM APIs to solve practical problems for our support, retention and operations teams - integrations, automation, agents and internal applications - rather than training or tuning models.
We are looking for solid software engineering fundamentals first. Experience with LLM APIs, AI-assisted development or the Salesforce platform is welcome and will help you move faster, and this role gives you the scope to build on it.
KEY RESPONSIBILITIES
Take ownership of locally-built reports, dashboards and automations, working with the people who built them to understand what the tool needs to do and what matters about it.
Replace manual steps with scheduled, automated processes so that nothing depends on a particular person being available.
Add the engineering that is usually missing - error handling, logging, alerting on failure, managed credentials, proper hosting and access control.
Document what you take on, so that it can be supported by someone other than its author.
Build and maintain integrations with the systems CE relies on, including CRM, Microsoft 365, Atlassian tooling, and our data warehousing and reporting platforms.
Work with REST APIs and authenticated services, handling pagination, rate limits, retries and partial failure sensibly.
Write reliable data pipelines - extracting, transforming and loading data on a schedule rather than on request.
Keep data handling consistent with Keyloop’s security and compliance requirements.
Use AI tooling - principally Claude, alongside Microsoft Copilot - as part of how you build, and apply it where it delivers real efficiency for CE.
Build integrations and small applications that use LLM APIs, with sensible attention to cost, reliability and whether the output is actually correct.
Develop enough familiarity with Keyloop’s internal support platform and its tooling to work alongside it and to assess what others have built on it.
Establish sensible defaults for how CE tooling is built, hosted, scheduled and documented, so that each new thing is cheaper to build and easier to support than the last.
Review and give an honest technical assessment of tooling built elsewhere in the function - what it does well, what it would take to support properly, and whether it duplicates something that already exists.
Use version control, testing and repeatable deployment as a matter of course.
Work directly with CE stakeholders - across Support, Retention, Operations and Knowledge & Education - to understand what they need and keep them informed.
Explain technical reality in plain business language, including when the answer is that something will take longer or should not be built.
Liaise with Keyloop’s wider Technology, IT and AI Centre of Excellence teams where your work touches theirs.
Making Existing Tooling Production-Ready
Integration & Data
Building with AI Tooling
Engineering Standards
Collaboration
Required Qualifications
Bachelor’s degree in Computer Science, Software Engineering, Information Technology or a related field, or equivalent practical experience.
3-5 years of professional software engineering experience, including building and running systems in production.
Strong TypeScript and Node.js, or strong Python with the ability to work confidently in TypeScript.
Proven experience integrating with REST APIs and authenticated third-party services - OAuth, token handling, pagination, rate limits, retries and error handling.
SQL, and experience working with data from more than one source system.
Experience automating manual or fragile processes - scheduled jobs, logging, monitoring and alerting when something fails.
Comfortable with containerised deployment, managed secrets, version control and repeatable release practices.
Able to work effectively in unfamiliar codebases, including code you did not write and would not have written that way.
Interest in AI-assisted development and LLM tooling, and a willingness to build with it day to day.
Self-directed, with the discipline to hold your own quality bar when nobody is checking - and to say clearly when something is not ready.
Able to communicate clearly in English, written and verbal, with stakeholders who are not technical.
Based in Ho Chi Minh City, or able to relocate there.
Desirable Qualifications
Salesforce experience- working with the Salesforce APIs and data model, Salesforce reporting, or Salesforce platform development. Useful, and not required.
Hands-on experience with LLM APIs - Claude, OpenAI or similar - including prompt design and evaluating output quality. Self-directed learning counts here as much as professional experience.
Familiarity with the Model Context Protocol (MCP), AI agent tooling, or Microsoft Copilot Studio.
Python, in addition to TypeScript.
Experience with data warehousing and BI tooling, such as Snowflake and Power BI.
Experience with Microsoft Graph, Atlassian (Jira and Confluence) or similar enterprise APIs.
Able to read PHP and work with a Symfony application, even without writing it day to day.
Background in or exposure to customer service, support or ITSM environments.
Experience in automotive, DMS or SaaS B2B software environments.
Experience working within a global delivery model, collaborating with teams in the UK or Europe.
Technical Skills
Category
Skills
Core languages
TypeScript, Node.js, JavaScript, SQL
Integration
REST APIs, OAuth 2.0, JSON, webhooks, error handling and retry patterns
Automation
Scheduled jobs, data pipelines, logging, monitoring, alerting
Deployment
Docker, containerised hosting, secrets management, Git, CI/CD
AI tooling
Claude, Claude Code, LLM APIs, prompt design, Microsoft Copilot
Desirable
Python, Salesforce APIs, MCP, Snowflake, Power BI, Microsoft Graph, Atlassian APIs, PHP/Symfony (reading)
Where This Role Grows
You will start by making existing CE tooling reliable, because that is where the immediate value is. As that work is completed, the balance shifts towards building new capability properly from the outset - deeper integration work, more ambitious automation, and more substantial use of AI agents and LLM-based tooling across the function.
It is also a genuinely visible role. What you build is used daily by the teams supporting 20,000 automotive retailers, and the difference between a tool that works and one that quietly breaks is felt immediately.
What Success Looks Like
Built an understanding of the CE function, the systems it depends on, and the tooling currently in use.
Taken ownership of at least one existing tool and made it run reliably without manual intervention.
Established working relationships with your key stakeholders across CE.
CE tooling that runs on a schedule, alerts when it fails, and continues working when individuals are unavailable.
Manual data handling steps progressively removed from CE processes.
Reliable integrations and well-documented, supportable solutions.
Consistent engineering standards across CE tooling, making each new piece of work faster to deliver.
Measurable reduction in time CE teams spend on manual reporting and data preparation.
In your first 90 days, you will have:
Ongoing measures of success include:
Why Keyloop
Work on real, high-impact problems in automotive - a sector undergoing rapid digital transformation.
Build the platform foundations behind a well-backed AI programme, with a clear path into AI development if you want it.
Join a collaborative, globally-connected CE organisation with leadership invested in making technology work for customers.
Access instructor-led Salesforce training and certification pathways through our Salesforce partnership, including Agentforce.
Grow your career within a company backed by Francisco Partners, with ambitious global expansion plans.
Competitive salary, flexible working, and a culture built around Experience-First - for employees as much as customers.