Senior Applied AI Engineer - Production AI at Scale

Groundswell; Connecticut; On-site; USD 150,000 - 205,000; Full time

Benefits offered by this job

Healthcare plans (medical, dental, and

FSAs (Flexible Spending Account)

401K match

Paid time off

Tuition & certs reimbursement

Flexible work schedule

On-site gym and childcare

Job summary

Groundswell is seeking an experienced AI/ML engineer to turn ambiguous business problems into working AI solutions in a client-facing, production environment. You will own end-to-end delivery from discovery through deployment, balancing accuracy, cost, latency, and security while mentoring others and shaping the roadmap.

You will work across client delivery, internal product development, rapid POCs, and enablement, collaborating with engineers, executives, and end users to produce robust AI

Qualifications

• 7+ years building and shipping production software.
• 2+ years hands-on experience building apps that integrate large language models, including prompt engineering, retrieval-augmented generation, structured extraction, tool use, and agent patterns.
• Demonstrated experience evaluating AI system quality, including building test sets, defining metrics, and making deployment decisions based on evidence.
• Strong programming ability in Python/TypeScript/SQL, with API integration and handling imperfect data across the full app, not just the AI layer.
• Proven ability to lead requirements conversations with non-technical stakeholders and translate to a technical approach.
• Sound judgment on accuracy, cost, latency, and complexity, with ability to explain tradeoffs to engineers and executives.
• Excellent written communication for client-facing documentation and design rationale.
• Ability to work with minimal direction in ambiguous situations, surfacing problems and driving decisions.
• U.S Citizenship and ability to obtain security clearance for client engagements.

Responsibilities

• Lead discovery with stakeholders to understand workflow, decision, and success criteria.
• Define measurable success criteria before building, including accuracy targets and acceptance conditions.
• Recommend against AI when simpler solutions suffice; consider rules, process changes, and better interfaces.
• Select the AI pattern appropriate to the problem (extraction, classification, summarization, retrieval, or agent).
• Defend architecture decisions on where workloads run considering quality, cost, latency, security, and authorization.
• Build the complete capability (prompts, retrieval design, APIs, data handling, UI) beyond AI components.
• Develop evaluation sets from real data and iterate on results rather than intuition.
• Determine when capability is ready for deployment and governance requirements.
• Operationalize capabilities with governance, logging, and traceability.
• Produce documentation detailing function, limitations, validation, and failure handling.
• Monitor quality, cost, latency, and drift post-launch and optimize continuously.
• Set direction and shape roadmap with working proofs, beyond presentations.
• Mentor engineers to raise the team’s judgment and output.
• Contribute reusable patterns to avoid starting from scratch.
• Move between client delivery, internal product work, POC, and enablement as needed.

Skills

Production software

LLM integration

Prompt engineering

API integration

Stakeholder communication

Programming: Python/TypeScript/SQL

AI project leadership

U.S. Citizenship

Leadership in AI projects

Tradeoff judgment

Education

Master's degree (Data Science/CS/Math)

Tools

Appian

OutSystems

Mendix

Power Platform

ServiceNow

Salesforce

Job description

Groundswell is seeking an experienced AI/ML engineer to turn ambiguous business problems into working AI solutions in a client-facing, production environment. You will own end-to-end delivery from discovery through deployment, balancing accuracy, cost, latency, and security while mentoring others and shaping the roadmap.

You will work across client delivery, internal product development, rapid POCs, and enablement, collaborating with engineers, executives, and end users to produce robust AI