Renata Baldissara-Kunnela AI Developer - LLM & Agent Systems Available for opportunities ================================================================ CONTACT ================================================================ Email: renatbk.linkedin@gmail.com Phone: +358 40 182 9008 LinkedIn: https://www.linkedin.com/in/renata-kunnela-4002a726a/ GitHub: https://github.com/RBKunnela Portfolio: https://renata-resume.vercel.app/ Location: Viitasaari, Finland ================================================================ SUMMARY ================================================================ AI Developer and AI Quality Engineering specialist with 15+ years in enterprise systems (banking, insurance, integration-heavy environments). I build production LLM and agent solutions embedded into business workflows using structured prompting, RAG grounding, fine-tuning, and tool orchestration. I also design reliability systems for modern AI: guardrails, fallback behavior, and automated evaluation (LLM-as-judge plus rule-based checks) used for regression testing and continuous improvement. ================================================================ STRENGTHS ================================================================ - End-to-end AI ownership. From prompt design and RAG grounding through guardrails, evaluation, and production reliability - full LLM lifecycle, not just the prototype layer. - Enterprise discipline applied to AI. 15+ years in banking and insurance environments under audit and compliance constraints; reliability and traceability as baseline, not bonus. - Cross-functional communication. Multilingual (Portuguese, Finnish, English, Spanish) and experienced bridging developers, architects, and business stakeholders across multi-system enterprise flows. ================================================================ KEY ACHIEVEMENTS ================================================================ - 4 AI products in production - including FriendlyAI assistants, Lumanela.fi (100-500 conversations/month), and Jurevo multi-agent orchestration. - 3 open-source frameworks published - ALMA-memory (PyPI + npm), AGtestari, and VERITAS-framework - all available on GitHub. - Conference speaker - Testaus ja AI 2024 (Finland's testing and AI conference). - 15+ years across regulated enterprise environments - banking- grade systems at Samlink and insurance integrations at TietoEvry/Fennia. ================================================================ EXPERIENCE ================================================================ FriendlyAI - AI Developer & Quality Engineering 02/2025 - Present - Built production AI assistants and automation flows for customer support and lead generation. - Implemented tracking and quality metrics across LLM-driven flows to make assistant behavior measurable. - Designed evaluation loops (LLM-as-judge plus rule-based checks) driving continuous improvement of prompts and responses. - Integrated LLM behavior into existing business workflows with focus on reliability, fallback paths, and safe escalation. TietoEvry / Fennia - API Tester & Quality Engineering 09/2023 - 02/2025 - Owned end-to-end API quality for RAML/MuleSoft APIs across complex insurance integrations. - Managed test cases and release evidence in TestRail and Jira; built exportable evidence packs for release-readiness reviews. - Built Postman test automation suites covering contract, integration, and regression scenarios. - Collaborated with developers and architects to resolve integration defects across multi-system insurance flows. Samlink Oyj - Senior Test Automation Engineer 01/2016 - 03/2023 - Developed scalable test automation for banking-grade systems using Robot Framework and Python. - Established QA practices for integration-heavy service chains under strict reliability and audit requirements. - Designed regression and release-readiness pipelines for production deployments in regulated environments. - Mentored testers and supported requirement clarification with developers and business analysts. ================================================================ SKILLS ================================================================ AI & LLM: OpenAI, Anthropic Claude, Azure OpenAI, Google Gemini, LangChain, LlamaIndex, RAG, Fine-tuning, Guardrails, LLM-as-judge, Prompt Engineering, MCP Engineering: Python, FastAPI, Pydantic, PostgreSQL, Redis, REST APIs, OAuth2, Docker, CI/CD QA & Testing: Robot Framework, Postman, TestRail, Jira, Confluence, RAML/MuleSoft, ISTQB Focus Areas: LLM & NLP (structured prompting, RAG grounding, fine-tuning, hallucination reduction, uncertainty handling); AI Agents (multi-agent orchestration, protocol-driven behavior, controlled tool access and permissions); AI Reliability (guardrails, prompt injection awareness, fallback strategies, safe escalation paths); Evaluation & Testing (LLM-as-judge plus rule checks, quality metrics, regression testing). ================================================================ LANGUAGES ================================================================ Portuguese - Native English - Excellent Finnish - Fluent (working language) Spanish - Fluent ================================================================ PROJECTS ================================================================ ALMA-memory (Open Source, PyPI, npm) Persistent memory layer for AI agents. Alternative to Mem0 with scoped learning, anti-patterns, multi-agent sharing, and MCP integration. Stack: Python, SQLite, MCP, FastAPI, npm Repository: https://github.com/RBKunnela/ALMA-memory AGtestari (Open Source, SaaS) AI-powered unified testing platform deploying autonomous QA agents into customer Azure environments. Consolidates frontend, backend, mobile, accessibility, and performance testing through a 12-phase enterprise workflow. Stack: FastAPI, React, Azure, AI Agents, Jira Repository: https://github.com/RBKunnela/AGtestari VERITAS-framework (Open Source) Evaluation framework for AI systems with LLM-as-judge capabilities, quality metrics, and regression testing support. Stack: Python, LLM-as-judge, Evaluation Repository: https://github.com/RBKunnela/veritas-framework Jurevo (Confidential, AI Architect) Multi-agent orchestration with controlled tools and a chat evaluation harness (LLM-as-judge plus rule checks). Modular backend and high-performance enterprise search. Stack: FastAPI, PostgreSQL, Redis, Multi-agent Lumanela.fi (Production) LLM with RAG and fine-tuning. Structured prompt logic and edge-function orchestration in production. Stack: LLM, RAG, Fine-tuning, Edge Functions ================================================================ EDUCATION ================================================================ BSc in Computer Science / Business IT JAMK University of Applied Sciences 2011 - 2016 MBA, International Business and Law Abo Akademi 2004 - 2006 BBA, Management Faculdade Candido Mendes, Brazil 1985 - 1990 Guest Lecturer JAMK University of Applied Sciences 2007 - 2016 ================================================================ CERTIFICATIONS ================================================================ ISTQB Foundation - Certified Tester 01/2014 ISTQB Foundation - Agile Tester 04/2015 Advanced Robot Framework 11/2015 Multimodal Prompting with Google Gemini 05/2024 Prompt Engineering Basics - Azure OpenAI 05/2024 Speaker - Testaus ja AI 2024 ================================================================ END OF RESUME ================================================================ Last updated: 2026-05-15 Canonical URL: https://renata-resume.vercel.app/resume.txt JSON version: https://renata-resume.vercel.app/resume.json