Andrés Lista
Full-Stack Engineer AI Training-Data Specialist
I annotate and evaluate the data frontier models learn from, and I build software. The overlap is where I'm most useful.
Florida
- Years in AI training data
- 4+
- Annotation & RLHF programs
- 4
- Bilingual evaluation
- ES / EN
Profile
About
I work at the intersection of AI training data and software engineering. Over four years and four annotation and RLHF programs — audio, image, and text — I have labeled and evaluated data for frontier-model and consumer-technology clients.
The engineering background is the practical difference. I can talk to technical staff directly about annotation tooling, schema design, and validation, instead of filing a ticket and waiting.
Track record
Experience
AI training & annotation
Frontier-model, consumer, social, and search clients
2026 — Present · Remote · Contract
AI Expert Evaluator
Invisible Technologies
Model-training tasks and rating criteria for a major search-technology client.
- Completed model-training tasks to project specification.
- Designed and applied custom rating criteria aligned to product objectives.
- Collaborated with cross-functional teams on model improvement and training-data curation.
- Rating criteria
- Data curation
- Model evaluation
January 2026 — July 2026 · Remote · Contract
AI Data Annotator & Trainer
RWS
Natural-language annotation for model training and fine-tuning at a large social-technology client.
- Annotated complex natural-language datasets for model training and fine-tuning.
- Validated technical and specialized content across industries.
- Participated in quality-control audits, tracing labeling inconsistencies to under-specified guidelines and proposing rule clarifications.
- NLP annotation
- Fine-tuning data
- QC audits
2024 — 2025 · Remote · Contract
AI Trainer & Rater
TELUS International
Annotation and preference rating for a frontier AI lab, across audio, image, and text-response data.
- Annotate audio, image, and text-response data for a frontier AI lab, applying project-specific rubrics across 1,000+ items.
- Rate and compare LLM outputs, identifying hallucinations, instruction-compliance gaps, and factual errors against custom scoring frameworks.
- Detect and document bias, toxicity, and safety issues, and escalate edge cases the guidelines do not cover so the ruling becomes spec rather than individual judgment.
- Rubric design
- Comparative rating
- Safety evaluation
- Audio annotation
2022 — 2023 · Remote · Contract
AI Quality Rater — RLHF
Welocalize
RLHF audio, image, and content rating for a major consumer-technology client.
- Evaluated multilingual Spanish and English responses for linguistic accuracy, cultural appropriateness, and context, including dialect-sensitive calls where regional variation is a feature and not an error.
- Produced written feedback reports on model outputs and guideline gaps to support improvement cycles.
- Ran validation testing on post-training datasets against defined quality thresholds, within AI safety and alignment evaluation frameworks.
- RLHF
- Multilingual evaluation
- Dataset validation
- Alignment review
Engineering
Full-stack product work, running alongside the annotation track
2025 — Present · Remote
Full-Stack Web Developer
TDBIM Solutions LLC
Full-stack application development on React and AWS.
- Build full-stack applications in React, TypeScript, Node.js, MongoDB, and AWS.
- Implemented automated validation and testing, reducing manual QA effort by 45% — the same defect-catching approach that keeps malformed annotations out of a training set.
- React
- TypeScript
- Node.js
- MongoDB
- AWS
2023 — 2024 · Hybrid
Full-Stack Web Developer
LESSA Investments LLC
React dashboards with real-time data processing and financial modeling.
- Built React dashboards with real-time data processing and financial modeling.
- Integrated the Azure Document Intelligence API, cutting document processing time.
- React
- Azure Document Intelligence
- Financial modeling
Toolkit
Skills & competencies
Annotation & evaluation
- Audio & speech annotation
- Multilingual evaluation (ES/EN)
- Transcription quality & consistency
- Dialect & accent variation
- Prosodic annotation
- Comparative rating
- Rubric fidelity
- Bias & safety detection
- Annotation schema design
AI & ML
- AWS Bedrock
- SageMaker
- Transcribe (ASR)
- Polly (TTS)
- LangChain
- Prompt engineering
- RLHF workflows
Languages
- Python
- JavaScript
- TypeScript
- Java
Stack & tools
- React
- Next.js
- Node.js
- FastAPI
- MongoDB
- PostgreSQL
- DynamoDB
- AWS
- Docker
- Git
- REST APIs
- CI/CD
Languages
- Spanish · Native
- English · Professional working proficiency
Get in touch
Let's talk
I reply to email within 24 hours. Write if you have an annotation program, an engineering role, or a question about training-data quality.