Role 1 - AI Software Validation Consultant

• Role 1 AI Software Validation Consultant
• Number of openings 1
• Location - Tampa, FL
  • Fully onsite
• Duration - 6 months

Job Title - AI Software Validation Engineer
Job Description
Experience Required: Prior hands-on experience validating software systems with embedded AI/ML
components in a GxP/pharmaceutical or biotech environment strongly preferred.
Role Overview
Support the qualification and software validation of two newly purchased AI-based automated plate
counters for a microbiology laboratory, as part of a broader FDA Warning Letter remediation program.
This is the client's first-ever validation of an AI-embedded software system, so the role requires the
ability to build a validation framework largely from first principles, adapted from precedent
methodology used elsewhere in the client's organization.

Key Responsibilities
Lead/support full lifecycle qualification (IQ/OQ/PQ) of two AI-based automated plate counters,
which are not yet installed, targeting a compressed October December delivery window.
Develop initial AI training and qualification protocols covering the model's base pre-training plus
site-specific training on the client's own plate types and labeling conventions.
Design and document a retraining/requalification framework defining triggers, frequency, and
verification steps for ongoing AI model maintenance post-qualification.
Design and execute comparative testing between AI-generated colony counts and independent
human reads, supporting the client's new 200% EM plate inspection model (AI as first 100%
read, human as second independent 100% read).
Design and execute edge-case/robustness testing to confirm the AI does not misread
ambiguous, overlapping, or atypical colony presentations.
Review vendor-supplied qualification documentation (noting it is currently high-level/basic) and
identify gaps requiring additional vendor technical documentation on the underlying AI model.
Support/recommend a supplier audit of the AI plate reader vendor, evaluating the vendor's
model development lifecycle controls (not just deployment-side controls) - proposed as an
optional scope item.
Assess data criticality: outputs feed directly into batch release records and can trigger
investigations for out-of-limits results, so validation must address data integrity requirements
for a locally installed (non-cloud) system.
Align validation approach with the client's existing AI software validation lifecycle methodology
(previously implemented at another client site), adapting for the technical differences specific to
this AI module.
Incorporate relevant regulatory/industry frameworks where applicable, e.g., USP < 1223>
(Validation of Alternative Microbiological Methods), and stay current on regulatory findings
trends affecting AI-based lab instrumentation.
Prepare documentation designed to withstand regulatory audit scrutiny, consistent with the
broader FDA Warning Letter remediation context.

Collaborate with the client's engineering, quality, and microbiology teams, and coordinate with
PQE's CSV and CQV teams supporting parallel workstreams (e.g., Filling Line 17).

Required Qualifications
Demonstrated experience validating software with embedded AI/ML components in a regulated
(GxP) environment; experience with automated plate readers/colony counters or similar lab
instrumentation is a strong plus.
Familiarity with computer system validation (CSV) principles and their application to AI-based
systems (e.g., model training/retraining governance, model drift considerations).
Experience designing comparative/parallel testing protocols (system vs. human performance)
and edge-case/challenge testing.
Working knowledge of data integrity requirements for locally installed (on-premises) GxP
software systems.
Familiarity with vendor qualification/audit practices, including assessing a vendor's
software/model development lifecycle controls.
Knowledge of relevant microbiological method validation standards (e.g., USP < 1223>) is a plus.
Strong technical writing skills for protocol and report development capable of withstanding
regulatory audit review.
Ability to operate effectively under accelerated project timelines.
Bachelor's degree in Life Sciences, Engineering, Computer Science, or related field (or equivalent
experience).