About Me
Engineer, maker, and lifelong learner
I'm a Computer Science Engineer from Universidad del Valle de Guatemala focused on building scalable systems at the intersection of backend architecture and applied AI. My work combines strong formal foundations with hands-on experience delivering production-ready full-stack solutions.
Currently working as a Fullstack Software Engineer at Nimble, building and extending a core Django/SvelteKit platform to deliver custom web applications across diverse client domains. My work spans REST API design, data modeling, responsive SvelteKit UIs, and third-party integrations — all under agile SCRUM workflows.
AI Specialization
Applied AI/ML track within my Computer Science Engineering degree at UVG
NLP & Language Models
- Transformer architectures: self-attention, positional encoding, encoder/decoder taxonomy (BERT, GPT)
- RAG pipelines: chunking, vector retrieval, and re-ranking to mitigate hallucinations
- Agentic systems: state-graph orchestration (LangGraph/ReAct) for planning and tool use
- Local inference, quantization & LoRA fine-tuning with Unsloth, Ollama and vLLM
- Evaluation via LLM-as-a-Judge and Ragas
ML Engineering (MLOps)
- From notebooks to production: modular Python, Sklearn pipelines, source/test/config separation
- Model serving: FastAPI/Flask APIs, Docker & Docker Compose packaging
- CI/CD for training and deployment with GitHub Actions and Azure DevOps
- Testing & validation: pytest, smoke tests, data validation with Great Expectations
- Production monitoring: drift, performance and resource usage with Evidently, Grafana, Prometheus
Computer Vision
- Classical vision: filters, convolutions, feature extraction, RANSAC & homographies for stitching
- CNN architectures (AlexNet, VGG, ResNet) with transfer learning and data augmentation
- Real-time detection & segmentation: YOLO, U-Net, object tracking
- Vision Transformers (ViT) and generative vision models: autoencoders, GANs, diffusion (Stable Diffusion)
Reinforcement Learning
- Formal foundations: MDPs, multi-armed bandits, dynamic programming, Q-Learning/SARSA
- Deep RL: DQN, Double/Dueling DQN, Actor-Critic, SAC, PPO, offline RL
- Multi-agent RL: CTDE, QMIX for cooperative and competitive environments
- Emerging directions: Decision Transformers, RLHF and LLM tool-use agents, CleanRL
Responsible AI
- Privacy-by-design: NIST AI RMF, PII handling, on-premise/local RAG to prevent data leakage
- Explainability: SHAP, LIME, PDP, Anchors — fairness evaluation and bias mitigation (Fairlearn, AIF360)
- Model documentation (Model Cards), red teaming and prompt-injection prevention
- FinOps for AI: token/API cost estimation, caching and rate-limiting strategies, latency and carbon footprint
Experience
Where I have applied my skills professionally
Fullstack Software Engineer
Building and extending a core Django/SvelteKit platform to deliver custom web applications for diverse clients, spanning backend architecture, REST APIs, and responsive frontend interfaces.
- Architect core full-stack platforms using Django REST Framework, SvelteKit, and TypeScript
- Engineer MLOps pipelines leveraging LLMs, Whisper, document processing, and RAG architectures
- Build responsive UIs with SvelteKit, TypeScript, and Tailwind CSS following pixel-perfect design specs
- Integrate third-party services including payment gateways, social auth, and email providers
- Work under SCRUM with 2-week sprints, participating from planning through production deployment
- Enforce software quality with automated pytest and Playwright suites, Docker containerization, and CI/CD
Junior Full-Stack Developer
Contributing to the development and evolution of a scalable payroll management platform, focusing on modern frontend architecture and reliable backend services.
- Migrated legacy MVP to a production-ready enterprise payroll platform
- Built RESTful services with NestJS, Node.js, TypeScript, and MongoDB
- Developed and maintained core features using React, Next.js, and MUI
- Collaborated with cross-functional teams to enhance UI/UX and ensure reliable production deployments
- Contributed to architectural improvements to support system growth and long-term maintainability
Skills
Technologies and tools I work with
AI & MLOps Engineering
Frontend Architecture
Backend & Systems
Languages
Cloud & MLOps Infrastructure
Systems & Networking
Projects
A selection of what I have built
CCBank
Cloud-native banking microservices deployed on AWS with full IaC infrastructure.
Education
My academic background
B.S. in Computer Science & IT Engineering
Universidad del Valle de Guatemala
Comprehensive computer science program with strong foundations in algorithms, operating systems, compilers, databases, networking, artificial intelligence, and software engineering.
- Advanced coursework in Compilers, Operating Systems, Parallel & Distributed Computing, and Deep Learning
- Built a complete compiler pipeline including lexical analysis, parsing, semantic analysis, and code generation
- Designed simulation and optimization models applying formal mathematical methods
Let's Connect
I'm always open to discussing new projects, creative ideas, or opportunities.