New opportunity

AI Solutions Lead

Omnea · New York, United States

About this role

Omnea’s AI Solutions Lead owns enterprise deployments of its procurement platform from discovery through launch and adoption, typically managing 3–5 deployments concurrently. The role combines customer-facing implementation with designing and validating AI-assisted configurations and the specialized agents customers use in their workflows. Success is measured through time to launch, post-go-live customer satisfaction and customer references.

Responsibilities

  • Discover requirements with Procurement, Finance, Legal and IT; design approval chains, routing, integrations and a realistic go-live plan with the customer sponsor.
  • Sequence deployments involving a dozen or more integrations, resolving source-of-truth decisions and record-reconciliation issues across systems.
  • Specify and configure intake and risk agents, including how agents pass structured requests and context through to approvals. Direct AI-generated first drafts of configurations and integration mappings, then validate outputs against customer policies, processes and edge cases. Make judgment calls on ambiguous requirements and custom logic.
  • Present workflows, address stakeholder concerns and run multi-day onsite workshops across multiple workstreams. Manage timelines, blockers and alignment; monitor adoption and adapt configurations after launch.
  • Share customer insights with Product and Engineering during design, and experiment with evolving AI tools and approaches.

Qualifications

  • 4–8 years in a fast-paced environment such as a startup, scale-up, consulting, banking or a venture you built yourself. Client-facing experience, or the ability to excel in client-facing work, and a record of exceptional achievement are sought.
  • Ability to give AI tools clear, specific instructions and critically evaluate their output. Prior use of ChatGPT, Claude or Cursor is a strong plus.
  • Ability to reason about system architecture, data models, primary keys and integration sequencing without necessarily writing integration code. Strong communication, organization, ownership and cross-functional stakeholder management are important.

The position is full-time and hybrid in New York, with in-office work on Tuesdays, Wednesdays and Thursdays. Compensation is US $120K–$190K per year, based on leveling during interviews, plus equity. Omnea also offers geo-localized benefits.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

AI agentsAgent orchestrationAI output validationPrompt writingWorkflow configurationIntegration mappingSystem architectureData modelingProject managementStakeholder managementClient communicationProcurement processesChatGPTClaudeCursor

Your skill match

Checking your profile…

YOUR NEXT STEP

Get interview-ready for this role

A focused preparation guide, built around this job’s responsibilities and requirements.

✦ AI-generated guide
Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.
01

Understand the role

What matters most

  • Translate discovery into a workable design

    The role runs discovery across Procurement, Finance, Legal and IT, then designs approval chains, routing and edge cases. Preparation suggestion: practice turning conflicting process descriptions into a proposed workflow, with open decisions made explicit.

  • Sequence integrations and resolve data ownership

    The posting assigns ownership of sequencing deployments with a dozen-plus integrations, including source-of-truth and record-reconciliation decisions. Preparation suggestion: be ready to explain dependencies, primary-key mismatches and how you would work with technical owners without writing the integration code.

  • Specify and validate connected AI agents

    The role configures customer-facing intake and risk agents so structured requests and context flow into approvals. It also directs AI-generated configuration drafts and validates them against policy and edge cases. Preparation suggestion: outline an agent handoff and the checks you would apply before launch.

  • Own launch and post-launch outcomes

    The lead typically runs 3–5 deployments concurrently, agrees success criteria and a go-live path with each sponsor, clears blockers and tracks adoption. The posting names time to launch, post-go-live satisfaction and customer references as outcomes. Preparation suggestion: structure an example around decisions, milestones and results.

  • Win stakeholder buy-in and surface product insight

    The role presents workflows from analyst to CFO level, runs multi-day onsite workshops and shares customer insight with Product and Engineering during design. Preparation suggestion: show how you would document disputed decisions and distinguish a customer-specific configuration need from product feedback.

Likely challenges to think through

  • Speed versus policy correctness

    Preparation hypothesis: AI-generated drafts may accelerate setup while missing a regional approval rule or other consequential exception. Prepare a validation approach that tests representative policies and edge cases, and makes human sign-off decisions explicit before shipping.

  • Integration dependencies versus launch dates

    Preparation hypothesis: a realistic go-live plan may be hard to maintain when source systems disagree on identifiers or ownership. Practice drawing a dependency sequence, identifying what blocks launch, and explaining which decisions need customer or technical-owner resolution.

  • Competing stakeholders versus a usable workflow

    Preparation hypothesis: additional controls requested by one team could make intake or approvals harder for others to adopt. Prepare to facilitate a trade-off among control, usability and timing in a cross-functional workshop, then check adoption after launch.

  • Parallel delivery versus sustained attention

    Preparation hypothesis: 3–5 deployments at different stages may create competing demands for workshops, validation and blocker resolution. Prepare a prioritization method that makes critical-path risks, decision owners and next actions visible across projects.

03

Practice your answers

Technical practice

How would you turn discovery across Procurement, Finance, Legal and IT into a deployment design?

What to coverSuggested answer focus (inferred): Explain how you would capture each team's process, decision rights and exceptions, then resolve conflicting requirements before configuring workflows. Show how you would confirm the design with the customer sponsor.

How would you design an approval chain that works across regions?

What to coverSuggested answer focus (inferred): Cover routing rules, delegation of authority and exceptions, including the posting's example of a regional rule missed by an HQ-focused draft. Explain how you would check the flow against customer policy and get the relevant stakeholders' sign-off.

How would you sequence a deployment with more than a dozen integrations?

What to coverSuggested answer focus (inferred): Map which systems and data must be ready before dependent workflows can be tested or launched. Explain how you would expose blockers early and revise the sequence without losing a realistic go-live path.

How would you investigate records that fail to reconcile across two systems?

What to coverSuggested answer focus (inferred): Identify the entity, its primary keys in each system and the agreed source of truth before assessing the mapping. Describe how you would work with technical owners to isolate the mismatch and verify the correction, without assuming you must write integration code.

How would you specify an intake agent and a risk agent for a customer?

What to coverSuggested answer focus (inferred): Define what information the intake agent needs, when it should ask a follow-up question and which risks matter under that customer's policies. Distinguish customer-specific rules from assumptions that need human confirmation.

What would you specify at each handoff from intake through risk assessment to approvals?

What to coverSuggested answer focus (inferred): Describe the structured request and context each stage must receive and pass on so an approval is based on the right information. Address what should happen when required information is missing or an earlier output is ambiguous.

How would you instruct an AI tool to draft a workflow configuration or integration mapping?

What to coverSuggested answer focus (inferred): Give it explicit customer requirements, policy constraints, expected output and known exceptions rather than a vague build request. Separate what the tool can draft from decisions that require customer or human judgment.

How would you validate an AI-generated configuration before launch?

What to coverSuggested answer focus (inferred): Compare the draft with the customer's documented process and policy, then test normal paths and consequential edge cases. Explain how you would record discrepancies, resolve ambiguous requirements and check the revised output before it ships.

How would you define success criteria and a realistic go-live plan with a sponsor?

What to coverSuggested answer focus (inferred): Tie launch readiness to agreed workflows, integration dependencies, stakeholder decisions and outstanding blockers. Distinguish a fast launch from a sound one, and relate the plan to the posting's stated outcomes: time to launch and post-go-live satisfaction.

How would you evaluate a deployment after go-live?

What to coverSuggested answer focus (inferred): Look for evidence of adoption and whether configured workflows still fit the customer's evolving process. Explain how you would prioritize adjustments and share design-stage or field insights with Product and Engineering when they reveal a broader product issue.

Behavioural practice

Tell me about a time stakeholders disagreed on a workflow or decision.

What to coverSuggested answer focus (inferred): Use a real situation involving competing priorities, explain how you surfaced the underlying concerns and reached or escalated a decision. State the result and what changed for the stakeholders.

Tell me about a time you led a complex, cross-functional customer workshop.

What to coverSuggested answer focus (inferred): Describe the situation, how you structured discussion across workstreams and what you did when priorities conflicted. Give the resulting decisions, open actions and effect on the project.

Tell me about a time you managed several projects at different stages simultaneously.

What to coverSuggested answer focus (inferred): Set out the competing deadlines and blockers, then explain your prioritization and communication with sponsors or teams. Give concrete results, including any trade-offs or course corrections.

Tell me about a time you exceeded a client's expectations while keeping the work on track.

What to coverSuggested answer focus (inferred): Identify the client's need and the action you personally took beyond the expected scope, without inventing an achievement. Explain the outcome for the client and how you managed its effect on other commitments.

Tell me about a time you raised an early customer insight with a product or engineering team.

What to coverSuggested answer focus (inferred): Describe the customer situation, the evidence you brought and how you worked through any disagreement about its significance. State what decision or change followed, or what you learned if no change was made.

02

Sharpen your knowledge

Topics to prepare

  • Procurement approval logic

    The posting calls for approval chains, routing and policy edge cases, including regional delegation-of-authority rules. Revise by tracing a hypothetical purchase through Finance and Legal approvals, then explain how a regional exception changes the path.

  • Agent handoffs and context

    Omnea describes an intake agent passing structured requests to a risk agent and then into approvals. Sketch the information each stage needs, what it passes onward, and how missing context could affect the next decision.

  • Enterprise data and integration dependencies

    The role requires reasoning about data models, primary keys, sources of truth and sequencing across a dozen-plus integrations, not writing integration code. For a hypothetical supplier record held in two systems, explain which decisions must precede mapping and reconciliation.

  • Deployment success and adoption

    The posting names time to launch, post-go-live customer satisfaction and customer references as outcomes, and calls for tracking adoption. Revise by defining a hypothetical go-live criterion and explaining what adoption evidence would prompt a configuration change.

  • Cross-functional decision-making

    The role involves competing priorities across Procurement, Finance, Legal, IT and other stakeholders. Examine a hypothetical request for an extra Legal sign-off: identify the policy question, workflow impact and decision that needs sponsor alignment.

Skills to revise

  • Turn discovery into a workflow

    Practice asking distinct discovery questions for Procurement, Finance, Legal and IT, then turn the answers into a draft approval chain with routing exceptions. This mirrors the posting’s discovery and configuration work.

  • Specify a chained agent workflow

    Write a sample specification for an intake agent’s follow-up questions, its structured handoff to a risk agent, and the context approvals receive. This practices the precise, cross-agent instructions the posting emphasizes.

  • Review an AI-generated first draft

    Using a hypothetical approval policy, draft a configuration with an AI tool if available, then check it against the policy and an omitted regional exception. The posting prioritizes directing first drafts and catching errors before launch; prior use of ChatGPT, Claude or Cursor is a plus, not a stated prerequisite.

  • Plan integration sequencing

    Create a dependency list for a hypothetical multi-system deployment, flag a primary-key mismatch and describe how records would reconcile. This practices the integration judgment the role requires without assuming coding duties.

  • Facilitate a rollout decision

    Rehearse a short workflow presentation to a skeptical Finance stakeholder and Legal stakeholder; close with the decision, owner and effect on go-live. The posting calls for stakeholder buy-in, multi-day workshops and active management of blockers.

04

Your preparation checklist

  • Prepare an end-to-end deployment walkthrough

    Choose a real project, if available, and outline discovery, decisions, launch, and adoption. Quantify outcomes you can substantiate; Omnea measures time to launch, post-go-live satisfaction, and customer references.

  • Draft a procurement workflow design

    For a clearly labeled hypothetical customer, map discovery questions for Procurement, Finance, Legal, and IT into approval chains, routing rules, edge cases, and sponsor-agreed go-live criteria.

  • Map integration dependencies

    Sketch a sample sequence for a dozen-plus systems. Show which integrations block others, where each entity’s source of truth sits, and how you would reconcile records with different primary keys.

  • Specify a chain of customer-facing agents

    Write a sample intake-agent specification with follow-up questions and structured output, then define what the risk agent receives and what context passes into approvals. Treat this as a preparation exercise, not an Omnea configuration.

  • Practice validating an AI-generated draft

    Prepare a clear instruction for an AI-drafted workflow or integration mapping and a review checklist covering customer policy, missing information, regional approval rules, and cases requiring human judgment.

  • Rehearse a stakeholder workshop and rollout plan

    Outline a multi-day workshop agenda and practice explaining a disputed approval flow to both operational and executive stakeholders. Add a simple tracker for timelines, blockers, and decisions across 3–5 concurrent deployments.

  • Prepare questions about delivery and feedback loops

    Ask how the team validates agent configurations before launch, monitors adoption and satisfaction afterward, and shares early customer findings with Product and Engineering during design.

AI-generated from the job information · Prepared Sep 30, 2026 (UTC)Useful preparation, not a guarantee of eligibility or an offer.

Availability last checked: Nov 4, 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in United States. Matched to this role, not your profile.

Explore more jobs
Finding similar opportunities…