Back to home

Daniel Hernandez

Computer Scientist · Payments & ML · Bogotá, Colombia (open to remote)

Summary

Computer scientist building payment integrations across Latin America. Background in production machine learning: computer vision, RAG systems, and data pipelines.

Experience

Software Engineer (Payments Team)
PayJoy
Aug 2025 – Present
Bogota, Colombia (Hybrid)
  • Integrate Bancolombia and Wompi payment rails handling 30K+ transactions/day, raising payment success rate from 90% to 99%.
  • Automate reconciliation for Bancolombia and Wompi settlements, cutting reconciliation time from 5–6 minutes to under 60 seconds per query (80%+ faster).
  • Build a 3-layer observability stack (Datadog APM, custom metrics, and alerting) and codify p95/p99 latency monitors, alerts, and dashboards in Terraform, accelerating incident detection.
  • Lead the Razorpay payment integration, enabling PayJoy's expansion into a new market.
  • Lead Bitso payment processor integration, projected to reduce annual transaction costs by ~$1M while improving reliability and scalability.
  • Engineer a zero-to-production Kasnet integration, one of Peru's largest payment networks, enabling QR-code and e-wallet payments and unlocking up to $3M in projected annual savings.
  • Roll out offline code-generation workflows across engineering, product, and operations, mitigating regulatory risk in Brazil.
Senior AI and Backend Engineer
Reshut Tech
Dec 2024 – May 2025
Miami, USA (Remote)
  • Led computer vision development for YOLOv7 and Detectron2 production models, improving accuracy and reliability through systematic fine-tuning and evaluation.
  • Architected GCP-based ML deployment workflows covering containerization, CI/CD, and auto-scaling, enabling consistent and repeatable production releases.
  • Introduced production monitoring and metric-based alerting for AI services, improving runtime reliability and cutting issue triage time.
Co-founder & Data Scientist
Dataplicada
Aug 2020 – Nov 2024
Miami, USA (Remote)
  • Built and deployed a RAG pipeline (LangChain, Llama3, GPT-4) that transformed large-scale customer feedback into actionable business insights.
  • Built Feedbacker from scratch, an end-to-end customer feedback product covering collection, analysis, and action.
  • Built Python ETL pipelines handling structured and unstructured data, accelerating data readiness for downstream analysis.
  • Scaled analytics platform to serve 100+ enterprise clients, enabling faster data-driven decision-making.
Software Engineer Intern
Tesla, Inc.
Sep 2022 – Dec 2022 (Concurrent Internship)
Reno, NV
  • Cut adhesive dispense waste by 20% by engineering real-time Python ETL pipelines that pinpointed Powerwall manufacturing bottlenecks.
  • Designed and built a predictive-maintenance system that forecasts Powerwall station failures from first-pass yield (FPY), downtime, and production telemetry, flagging at-risk stations before they go down.
  • Built Tableau dashboards and Python visualizations on Ignition SCADA data, giving production engineers clear visibility into Powerwall line performance.
  • Hardened assembly-station logic and data-integrity checks across 4+ Powerwall configurations, eliminating serial-number conflicts.
  • Established Kubernetes-based CI/CD with Docker and Jenkins, automating build, test, and release stages.
Software Engineering Consultant (Toyota Chile, Ford Argentina)
Independent Consultant
Jan 2018 – Dec 2020
Remote
  • Delivered a production VR showcase for Toyota Chile used in live customer-facing marketing campaigns.
  • Integrated Ford Argentina APIs into a multi-dealership CRM, improving vehicle-order tracking accuracy and reliability across sales teams.

Technical Skills

Python, SQL, C++, PyTorch, XGBoost, Computer Vision, RAG, FastAPI, GCP, Terraform, Datadog APM, CI/CD, Docker, Kubernetes, ETL, Tableau.