Services

Project-based engagements. Clear deliverables. Aligned incentives.

I work on projects, not hours. Each engagement has defined scope, deliverables, and success criteria agreed upfront. Hourly billing incentivizes slow work. Fixed-scope projects align my interests with yours—I'm motivated to solve the problem efficiently, not to maximize billable time.

I take on a limited number of projects at a time to ensure full attention and quality. If timing doesn't work, I'm happy to recommend trusted colleagues or revisit in a future quarter.

Engagement Types

Technical Assessment

2-4 weeks | Fixed scope

Deep-dive into your existing ML system, AI strategy, or technical architecture. Independent analysis of what's working, what's not, and what to prioritize.

Deliverables:
  • Written report with findings and prioritized recommendations
  • Architecture review or system audit documentation
  • Executive summary for leadership
  • 90-minute walkthrough call
Best for: Companies with existing ML that's underperforming, pre-acquisition technical due diligence, or "should we build or buy" decisions.

Implementation

2-6 months | Milestone-based

End-to-end build of ML system from requirements through production deployment. I handle architecture, development, integration, and handoff to your team.

Deliverables:
  • Production-ready ML system
  • Technical documentation and runbooks
  • Team training and knowledge transfer
  • 30-day post-deployment support
Best for: Companies with a clear problem, defined success metrics, and need for senior execution without hiring full-time.

Advisory

Quarterly retainer | Ongoing

Fractional AI leadership for companies building internal capability. Architecture review, vendor evaluation, hiring support, and technical mentorship.

Includes:
  • Regular strategy sessions (bi-weekly or monthly)
  • Async access for technical questions
  • Architecture and code review
  • Quarterly roadmap planning
Best for: Companies with growing AI teams who need experienced guidance without a full-time executive hire.

Specializations

I focus on three areas where I have deep domain expertise and production experience:

Compliance ML—Adverse media screening, entity resolution, risk classification for AML/KYC. I understand both the technical requirements (NER, classification, entity linking) and the regulatory context.

Industrial Inspection—Signal processing and computer vision for NDT, defect detection, predictive maintenance. Experience with proprietary data formats and integration with engineering workflows.

Enterprise GenAI—RAG systems, LLM orchestration, agentic workflows. Focus on retrieval quality, chunking strategies, and production reliability—not demo magic.

Project Criteria

I'm selective about projects to ensure I can deliver real value. Here's what makes a good fit:

Success metrics
Clear, measurable definition of success (F1, precision/recall, business KPI). If we can't define what "working" means, we shouldn't start.
Decision-maker access
Direct line to someone who can make decisions. Projects stall when approval chains are long.
Timeline
Minimum 4 weeks for assessments, 2+ months for implementation. No "we need it yesterday" projects.
Data access
For ML projects: access to representative data before we commit to scope. No "we'll figure out data later."

What I Don't Do

To respect your time and mine:

  • Staff augmentation—I'm not a contractor to fill a seat. I take ownership of outcomes, not hours.
  • Projects without metrics—If we can't define success, we can't achieve it.
  • "Explore AI possibilities"—I solve specific problems, not run innovation theater.
  • RFP processes—Procurement-driven engagements optimize for the wrong things.
  • Investor deck AI—I build systems that work, not stories that raise money.

This isn't arrogance—it's pattern matching. These project types have low success rates. I'd rather spend my time on work that actually ships.

How It Works

1. Initial conversation—30-minute call to understand your problem, constraints, and goals. No pitch, no pressure. We figure out if there's a fit.

2. Scoping—For implementation projects, I often do a paid 1-week discovery sprint to understand the problem deeply before committing to full scope. This protects both of us from bad estimates.

3. Proposal—Fixed-price proposal with clear deliverables, timeline, and success criteria. Milestone-based payments for longer projects.

4. Execution—Regular check-ins, transparent progress updates, no surprises. You'll always know where the project stands.

5. Handoff—Documentation, knowledge transfer, and post-project support. The goal is for your team to own the system independently.

Start a Conversation

Have a specific problem in mind? Let's talk about whether I can help.

Email me with a brief description of your situation and what you're trying to achieve. I'll respond within 48 hours with either next steps or an honest "not a fit."

k.wikiel@gmail.com