Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m an AI evaluation and data analytics specialist based in Bristol, with a Master’s in Artificial Intelligence and Technology Management. Over the last couple of years, I’ve led analytics for a UK care-staffing agency—modeling payroll, workforce, and financial data in Python, SQL/PostgreSQL, and Power BI, then translating the results into clear reporting for stakeholders and investors. I also work as an AI evaluation contractor, assessing model-generated documents against fixed rubrics by checking both realism and structural intent, and grounding every judgment in written, quotation-backed rationales. Alongside evaluation, I’ve built and shipped production web applications (including a SaaS product) and I continue to bring the same “auditable evidence” mindset to both data work and software—carefully measuring what’s correct, what’s missing, and what truly reflects practitioner output.…I’m an AI evaluation and data analytics specialist based in Bristol, with a Master’s in Artificial Intelligence and Technology Management. Over the last couple of years, I’ve led analytics for a UK care-staffing agency—modeling payroll, workforce, and financial data in Python, SQL/PostgreSQL, and Power BI, then translating the results into clear reporting for stakeholders and investors. I also work as an AI evaluation contractor, assessing model-generated documents against fixed rubrics by checking both realism and structural intent, and grounding every judgment in written, quotation-backed rationales. Alongside evaluation, I’ve built and shipped production web applications (including a SaaS product) and I continue to bring the same “auditable evidence” mindset to both data work and software—carefully measuring what’s correct, what’s missing, and what truly reflects practitioner output.

Alim Bidmus

AI Evaluator, AI Trainer, AI Prompt Engineer, +6





I’m an AI evaluation and data analytics specialist based in Bristol, with a Master’s in Artificial Intelligence and Technology Management. Over the last couple of years, I’ve led analytics for a UK care-staffing agency—modeling payroll, workforce, and financial data in Python, SQL/PostgreSQL, and Power BI, then translating the results into clear reporting for stakeholders and investors. I also work as an AI evaluation contractor, assessing model-generated documents against fixed rubrics by checking both realism and structural intent, and grounding every judgment in written, quotation-backed rationales. Alongside evaluation, I’ve built and shipped production web applications (including a SaaS product) and I continue to bring the same “auditable evidence” mindset to both data work and software—carefully measuring what’s correct, what’s missing, and what truly reflects practitioner output.…I’m an AI evaluation and data analytics specialist based in Bristol, with a Master’s in Artificial Intelligence and Technology Management. Over the last couple of years, I’ve led analytics for a UK care-staffing agency—modeling payroll, workforce, and financial data in Python, SQL/PostgreSQL, and Power BI, then translating the results into clear reporting for stakeholders and investors. I also work as an AI evaluation contractor, assessing model-generated documents against fixed rubrics by checking both realism and structural intent, and grounding every judgment in written, quotation-backed rationales. Alongside evaluation, I’ve built and shipped production web applications (including a SaaS product) and I continue to bring the same “auditable evidence” mindset to both data work and software—carefully measuring what’s correct, what’s missing, and what truly reflects practitioner output.

Available to hire

I’m an AI evaluation and data analytics specialist based in Bristol, with a Master’s in Artificial Intelligence and Technology Management. Over the last couple of years, I’ve led analytics for a UK care-staffing agency—modeling payroll, workforce, and financial data in Python, SQL/PostgreSQL, and Power BI, then translating the results into clear reporting for stakeholders and investors.

I also work as an AI evaluation contractor, assessing model-generated documents against fixed rubrics by checking both realism and structural intent, and grounding every judgment in written, quotation-backed rationales. Alongside evaluation, I’ve built and shipped production web applications (including a SaaS product) and I continue to bring the same “auditable evidence” mindset to both data work and software—carefully measuring what’s correct, what’s missing, and what truly reflects practitioner output.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

AI Evaluation & Preference Rating Contractor at PocketClass Limited

January 1, 2025 - Present

Assesses AI-generated documents against reference material using fixed rubrics, judging whether content and structure reflect what domain practitioners would genuinely produce. Produces quotation-backed rationales for each judgment and grades batches point-by-point on a 1–7 scale. Applies forced score-distribution rules to ensure comparative discrimination, flags ties/ambiguities and underspecified references, and documents how frontier model failures appear when surface requirements are satisfied but structural intent is missing.

Ophthalmic Technician at University Hospitals Bristol & Weston NHS Foundation Trust (Bristol Eye Hospital)

January 1, 2024 - Present

Acquires and records OCT scans, IOP results, and Humphrey visual field reports into structured, audit-ready clinical records. Performs protocol-driven diagnostic assessments and maintains precision and consistency across high-volume encounters with documented reasoning suitable for clinical audit.

Senior AI Response Evaluator at Nexus AI Systems

January 1, 2024 - Present

Led a team of 15 evaluators, curating high-quality SFT datasets for a multi-modal LLM launch. Developed proprietary scoring rubrics for “Reasoning” and “Creative Writing,” improving inter-annotator agreement by 18%. Ran adversarial testing to identify and mitigate bias in response generation, strengthening response quality and safety.

Director at MedVendorHub — Supported Living Staffing & Care Agency

June 1, 2023 - January 31, 2026

Owned the agency’s data and reporting function, analyzing payroll, financial figures, and staff statistics using Python (pandas), Excel, SQL, and Power BI for reports delivered to stakeholders and care home investors. Wrote Python scripts to clean/reconcile/join payroll exports into analysis-ready datasets, maintained a PostgreSQL database for traceable personnel and compliance records, and ensured gaps/unverifiable entries were flagged. Led pre-placement due diligence (right-to-work verification, employment history/reference checks, and professional registration/clearance confirmation). Drafted contractual and statutory documentation including employment contracts aligned with UK employment and care-sector regulation.

Frontend & Full-Stack Developer at MedVendorHub (Independent) / Upwork / Fiverr

May 1, 2022 - December 31, 2025

Built and shipped production web interfaces using HTML/CSS and TypeScript, including structured data handling and JSON delivery against defined schemas. Built Repurpose Studio, a live full-stack SaaS product with users (Next.js/TypeScript/Supabase/Prisma/Stripe), owning the frontend end-to-end. Used browser DevTools daily to inspect rendered markup and network behavior, focusing on what pages actually do rather than appearances.

RLHF Content Specialist at OpenText Dynamics

January 1, 2022 - January 1, 2024

Ranked 10,000+ model comparisons using helpfulness, honesty, and harmlessness (HHH) criteria. Partnered with research engineers to troubleshoot edge-case hallucinations in technical coding responses. Authored style guides to support consistent model personality alignment across responses.