Senior AI Operations Manager
The Problem
Our AI gives inconsistent answers and our teams don't trust it for critical decisions. We waste hours fixing AI mistakes because it doesn't understand our business context. The models themselves are fine - it's how we're training them that's broken.
We feed data but don't shape judgment. We retrieve information but don't enforce thinking standards. We prompt for outputs but don't condition behavior. This leads to inconsistent decisions, brittle performance, and low trust.
We need someone who treats AI deployment like operator training, not software deployment. Someone who builds the discipline architecture that makes AI systems predictable, safe, and actually useful.
What We're Dealing With
EviSmart deploys AI across multiple internal systems: scan quality control, autonomous CAD design, HR automation, and business intelligence. These systems make thousands of decisions daily that affect production quality, customer satisfaction, and operational efficiency.
You'll own the methodology that makes our AI systems behave consistently, safely, and usefully. We call it Deliberate AI Conditioning (DAC): behavioral training through cognitive framing, constraints, scenario drills, and critique loops. Train AI like you'd train operators, not like you'd configure software.
You'll report to the CEO and partner with squad leads across AI QC, Autonomous CAD, HR Hand, and 2Brain to implement this methodology. You own the standard; they own the systems.
What You'll Actually Do
Cognitive framing:
- Define role identity for each AI system - what role is the AI playing? What does success actually look like?
- Establish tradeoff priorities - when priorities conflict, what wins?
- Create measurable success criteria that align AI behavior with business outcomes
Behavioral constraints:
- Build explicit rules with no ambiguity - no "it depends"
- Define hard stops, soft limits, escalation triggers, output formatting standards
- Create constraint libraries that can be reused across AI systems
Scenario drills and critique loops:
- Design edge case libraries - the weird situations that break AI judgment
- Build self-critique mechanisms - AI that checks its own reasoning before outputting
- Require 20+ documented failure scenarios before any AI system launches
Memory and reinforcement:
- Design systems that learn from corrections - not just data, but judgment patterns
- Build feedback loops that reinforce good behavior and flag drift
- Create the DAC Certification process for all AI deployments
What You Need
Requirements:
- Deep experience with LLM prompt engineering and behavioral conditioning
- Track record of deploying AI systems in production environments
- Ability to translate business logic into AI behavioral rules
- Systems thinking - you see how AI decisions cascade through operations
- Strong documentation skills - you'll be creating the playbooks others follow
Nice to have:
- Experience with RAG systems, vector databases, and knowledge retrieval
- Background in operations, quality control, or process engineering
- Experience training humans on complex procedures (you understand how learning actually works)
- Healthcare or dental industry experience
How This Works
This is hands-on. You'll work directly with our AI systems - writing prompts, testing edge cases, documenting failure modes, and building the training architecture that makes AI predictable.
Why This Matters
Most companies treat AI like magic - throw data at it and hope for the best. We're building the discipline architecture that makes AI actually work in production. You'll define that standard.
You'll work across multiple AI systems with real business impact - quality control, design automation, HR, business intelligence. This isn't a research role. It's operational AI at scale.
The dental industry is going through a massive digital transformation. We're at the center of it. This is infrastructure work - once we're embedded, we don't get ripped out.
EviSmart
www.evismart.com
Create a Job Alert
Interested in building your career at Evismart? Get future opportunities sent straight to your email.
Apply for this job
*
indicates a required field