# Juan Ignacio Castore Bergioli

> Product Engineer at a stealth AI startup · Co-founder ×3 · BSc in Digital Technologies (Major in Computer Science) @ UTDT, GPA 9.13/10 · Buenos Aires, Argentina

**Mission:** Change the world by building frontier technology. Push the line of what's possible.

- Website: https://juanicastore.com
- CV (PDF): https://juanicastore.com/Juan_Ignacio_Castore_CV.pdf
- LinkedIn: https://linkedin.com/in/juan-ignacio-castore
- GitHub: https://github.com/juani-castore
- Email: juanicastore.ftw@gmail.com

## About

Product engineer at a stealth AI startup automating companies' back-office operations: I own one of our two customer segments end-to-end — I interview users, define the roadmap, and write the code that ships it. I led the adoption of AI-native development across 100% of the company: 8.8× engineering velocity, measured week over week. Co-founder of three startups, BSc in Digital Technologies (Major in Computer Science) at UTDT with a 9.13 GPA, and a Teaching Assistant there since 2023. Before UTDT, I trained as a backend developer on a full scholarship funded by Mercado Libre and Globant. My long-term goal is simple to state and hard to do: build frontier technology that changes how the world works.

## Now (updated Jul 2026)

- Owner of the Accountants segment at a stealth AI startup — interviews with accounting firms, roadmap, and code
- WhatsApp receipt ingestion now in production: photo → extraction → validation → ERP — zero receipts lost since launch
- Leading an AI-native engineering culture: 8.8× measured dev velocity, internal target 15×
- CTO at Orion — agent automation for the insurance sector
- Teaching Assistant at UTDT: Algorithm Design, Computer Systems, Intro to Programming

## What drives me

- **Frontier technology** — The long-term north star: working where technology has no manual yet — AI and agentic systems today; quantum computing and graduate research on the horizon.
- **AI-native by conviction** — AI multiplies whoever has judgment. I make the architecture and product decisions, and build with agents to move faster and more robustly — 8.8× measured, not promised.
- **Owner of outcomes, not tasks** — What's scarce isn't knowing how to build: it's judgment + user contact + the ability to ship the decision. That's why I take on whole segments and products, from interview to deploy.

## Experience

### Product Engineer — Operational Agent & Accountants Segment Owner — Stealth AI Startup — OpsTech (back-office automation) (Sep. 2025 – Present)

- Lead of the Operational Agent: the AI agent that handles each client's operations directly against their own ERP. My concrete operational responsibility: no incoming receipt is lost, and uptime stays >99% for the 200+ organizations operating daily. I define product roadmap and features; I report directly to the CTO.
- Owner of the Accountants segment since Jun 2026 — 1 of the company's 2 customer segments: mandated to interview as many accounting firms as possible, understand their pain points, and evolve the product for them. The full cycle: user research → roadmap → code.
- Designed and built the WhatsApp ingestion channel from scratch — now in production: accountants photograph a physical receipt and it flows straight into the agent's pipeline (extraction → validation → ERP load), no manual digitization. A microservice I founded commit by commit (WhatsApp Business webhook → GCS → Cloud Tasks → ingestion), with reverse-OTP sender verification and an automatic reply to each message with the receipt's result. Not a single receipt has been lost since launch. The centerpiece of my Accountants-segment ownership. [0 receipts lost since launch]
- Picked by the CTO as the engineering reference to evaluate Claude Code: I mastered it in ~1 month, concluded it was worth adopting, and led its rollout while training the team. I scaled usage beyond engineering —operations, support, sales and leadership, including the CEO— to 100% of the company (~15 people): with Claude connected to the repos, the whole organization shares context and resolves technical and product questions autonomously. [100% company adoption (~15 people, all areas)]
- Within the first 2 months of the engineering team's adoption, the practices and workflows I proposed lifted development throughput to 8.8× the prior baseline —a metric we track week over week— against an internal target of 15×. [8.8× dev throughput vs. baseline (target 15×)]
- Foundational contributor to the product — my first feature shipped end-to-end (backend + frontend) in week one: 490 merged PRs and 53 production database migrations in 10 months, spanning the backend, the frontend and 3 ingestion microservices (mail, WhatsApp, document reading). [490 merged PRs · 53 DB migrations in 10 months]
- Led the 'never lose a receipt' program on the mail channel — the area's most serious problem when I took it over: redesigned processing atomicity, added persistent per-attachment tracking, a recovery mode, a visibility dashboard and an automated audit agent that runs weekly; also migrated the legacy ledger to the new system with dual-read gates and zero loss. The operational mandate is being met: user-visible lost receipts, zero. [User-visible lost receipts: 0]
- Designed and implemented the invoice auto-loading flow that eliminated manual form-filling for 40% of processed invoices, dropping user time to zero in those cases. [40% of processed invoices with zero user input]
- Stack I maintain: backend Python + FastAPI + SQLAlchemy 2 + Alembic on Cloud Run over Postgres; frontend Next.js 15 + React 19 + Shadcn + Clerk + React Hook Form/Zod; reading-service Python on Cloud Run with AWS Textract + OpenAI for structured invoice extraction, queued via Cloud Tasks. End-to-end logging and monitoring to sustain the uptime SLA.
- The company integrates 12 Argentine ERPs in production today. I led the hardest one — a legacy ERP with undocumented on-premise SOAP/XML — and deepened it into Argentine-market financial features (installments, split cash accounts, cheques, cost centers) verified in production; I contribute code to several adapters in addition to owning the product.
- Led the integration with ARCA (formerly AFIP, the Argentine tax authority): validation of CUIT, CAE, sales point and amount of each invoice against the agency before loading it into the ERP — a critical compliance step for operating in Argentina.
- Every feature on the Operational Agent starts with an explicit business case: resource cost vs. expected impact on the company AND on the end user. Scope is right-sized accordingly — from 2-day fixes to month-long builds — to avoid spending resources on solutions that don't move the metrics that matter.
- Closed-loop measurement discipline: every shipped feature is instrumented at the user level, and post-ship metrics feed back into the next prioritization round (what to double down on, what to deprecate, what to build next). 48 success metrics currently run under my name —each with an owner, a target and a history— and none has failed. The data loop is central to the decision process, not an add-on. [48 success metrics owned · 0 failed]
- Designed the multi-role invoice approval workflow (inspired by GitHub PRs and SAP release strategies) for enterprise expense control and evolved it to Phase 2 self-service: each organization configures its own routing rules from an editable UI, no code changes. Stack ~88% ready for pilot.
- Founded the company's knowledge-management system — it's now a team rule that every learning gets capitalized into a canonical doc: I wrote 63 learnings (the team's top producer) and built 12 internal tools plus an automated audit agent, bringing to an entire organization the knowledge-management vision I had originally conceived at Kora.

### Co-Founder & CTO — Orion Studio Automation (Feb. 2026 – Present) — https://automatizaciones-orion.com

- Co-founded Orion, an automation consultancy focused on the insurance sector. Single paying client to date (insurance broker, recurring pilot), in active conversations with two insurers to scale the solution across their broker networks.
- Designed and implemented a 5-tier entity resolution pipeline (canonical normalization, token subset/overlap analysis, Levenshtein fuzzy matching, LLM semantic validation) matching ~4,280 contacts against ~250 weekly policy renewals, with automated delivery via WhatsApp. Target match rate >90%.
- Self-hosted cloud architecture: Hetzner VPS CX23 (Ubuntu 24.04) with Docker Compose (n8n + PostgreSQL 16 + Redis + Evolution API / WhatsApp gateway), Cloudflare zero-trust tunnel, automated daily backups to Google Drive. Migrated from a prior on-premise deployment on a repurposed personal machine.
- As CTO, I own infrastructure, major implementations, and server operations; I keep the platform stable enough that my co-founder (CEO) can directly edit specific n8n nodes and handle client needs without blocking on me.
- First-class multi-tenant architecture: the same stack serves multiple brokers/insurers with logically separated data, ready to scale to full networks via insurer partnerships.
- FastAPI BFF (`orion-api`) exposing CSV/contact upload to the frontend and orchestrating n8n calls; auto-deploy to the VPS via GitHub Actions over SSH on every merge to main.
- FASE 2 in development: automatic capture of incoming WhatsApp payment receipts, stored in hierarchical filesystem YYYY/MM + Postgres, ready for posting to the insurer's portal.

### Co-Founder & CTO — Pickap (Feb. 2025 – Feb. 2026) — https://pickap.app

- Led full-stack development of a gastronomy ordering platform with scalable architecture (React, Node.js, Firebase).
- Designed and implemented RESTful APIs and integrations with Mercado Pago and Fudo.
- Managed the technical roadmap, real-user validation, and public product launch.
- Automated internal processes using Node.js and serverless architecture (Firebase Functions), improving operational efficiency.
- Built the 'kitchen ticket printer' as a Node.js service running on Raspberry Pi: real-time Firestore listener (`onSnapshot`) that prints each new order on a thermal printer and marks it as printed, ensuring no order is lost in the kitchen.

### Teaching Assistant — Universidad Torcuato Di Tella (Jul. 2023 – Present)

- Teach practical/lab sessions and design and grade assessments across three core Computer Science courses: Algorithm Design (algorithm design and analysis, complexity and theory of computation incl. NP-completeness, with its mathematical foundations; in Python and C++), Computer Systems (operating systems, architecture, C and Assembler), and Introduction to Programming (the university's foundational CS course, in Python). Code review at scale on hundreds of student submissions.
- Developed the ability to explain complex technical concepts clearly to diverse audiences — directly applicable to technical consulting and communicating with non-technical stakeholders.

## Selected projects

### contactarprofesionales — Local Services Marketplace (2026) — https://contactarprofesionales.com

- Independently built and deployed a marketplace connecting users with local service professionals in San Juan (plumbers, electricians, psychologists, etc.) with direct WhatsApp contact and event tracking.
- Stack: Next.js 15 + React 19 + Tailwind CSS 4 + Supabase (PostgreSQL + Auth + Storage), Server Components for SEO, RLS and Server Actions, auto-deploy to Vercel from GitHub.
- Live features: public directory with full-text search, self-service professional registration, department-level geolocation, WhatsApp contact with lead tracking.

### Kora — Digital Onboarding SaaS (Co-Founder) (2025) — https://koraonboarding.com

- Co-founded Kora in 2025: SaaS for digital onboarding automation. Today it runs with a dedicated team and my role is advisory — my active focus is on my main role, Orion and UTDT.
- Designed and implemented the RAG system with a dynamic knowledge base: plain-text files as user-editable memory, with an index system for efficient contextual access.
- Developed the user-editable 'Agent Memory' feature — direct context management and real-time customization of AI behavior.

### Conversational AI Agent for Soyana (Undergraduate Thesis) (2025)

- Developed an autonomous commercial conversational agent using AI (Gemini, Whisper) to automate order intake, orchestrated with n8n on GCP (139 nodes, 11 use cases).
- Specialized multi-agent architecture: Router Agent, Conversational Agent, Order Validator — each with a focused task to optimize debugging and prompt performance.
- Automated 11 critical use cases including order intake with disambiguation, volume discounts, order history queries, and human handoff.
- Measurable result: customer lookup time reduced from 4 min to 30 sec (8× improvement); monthly catalog and weekly reminders fully automated.

### bondiCounter — CABA Bus Tracking via QR (2024)

- Independent project with real traction: real-time tracking of buses in Buenos Aires (React + Leaflet + Node.js + public city transport APIs). Distributed physical QR codes on many bus stops so passengers could scan and see where their bus was.
- Got genuine scans from real passengers but the model wasn't profitable — full lifecycle learning: idea → MVP → physical distribution → real usage → kill due to unit economics.

### ADO-1 — Autonomous Digital Organism (Personal Research) (2026 – Present)

- Ongoing personal research: architecture of an autonomous digital organism (ADO-1) with an OODA loop (Observe-Orient-Decide-Act), four-tier cognition (S1 executor, S2 strategist, S3 supervisor, S4 override) and active guardrails against malicious patterns. Isolated in a Docker sandbox with `network_mode: none`, executes controlled actions on the host via PowerShell.
- Stack: Python + Docker + OpenAI API (with Ollama local fallback). Next step: release as open source.

## Additional projects

- **mialquiler.app — Property Management SaaS** (2025) — Independently built a full-stack SaaS application (PostgreSQL/Supabase) to automate administrative management between landlords and tenants.
- **Orbital Trajectory Optimization (Engineering Modeling)** (2025) — Solo project: parametric optimization pipeline for LEO orbits via massive grid search (10K+ candidates) with physical constraint validation (200–2000 km band, peak-to-peak radial variation, angular monotonicity).
- **Audio Deep Learning — Genre Classifier + Autoencoder** (2024) — End-to-end deep learning pipeline for audio in PyTorch (3-person group project, TDVI course): 2D CNN over mel-spectrograms with BatchNorm/Dropout/data augmentation and experiment tracking in Weights & Biases.
- **University Machine Learning Competition** (2024) — 2nd place. Built an ML pipeline with CatBoost as the final model, including feature engineering, exploratory analysis, and hyperparameter tuning.
- **Profitability Analysis — Wind Turbine in Patagonia (Monte Carlo)** (2024) — Consulting case for the course Computational Applications for Business (UTDT): Monte Carlo simulation to evaluate the profitability of installing wind turbines in Argentine Patagonia, modeling wind variability, energy price variability, and turbine failure risk as stochastic variables.
- **Schedule Optimization — Retiro–Tigre Train Line** (2024) — Algorithm Design coursework (UTDT — the course where I'm now a Teaching Assistant): formulated and solved a scheduling problem for the Retiro–Tigre train line, maximizing demand met over operating cost, modeling tradeoffs between frequency, capacity, and peak usage.
- **Facility Location — Combinatorial Optimization** (2024) — Optimization coursework in ACN (UTDT): optimal assignment of 30 clients to 10 telephone exchanges in greater Buenos Aires using a given distance matrix — classic facility location problem solved with constructive heuristics.
- **Coffee Market — Dynamic Pricing (Strategic Management)** (2025) — Capstone project for the Strategic Management elective (business minor, UTDT): Business Model Canvas for a coffee chain with stock-market-style dynamic pricing. Acted as CTO of the 5-person team: defined pricing strategy and the technical pipeline.
- **Deep Q-Learning Agent for Connect 4** (2025) — Group project for the Artificial Intelligence & Neuroscience course (UTDT): Deep Q-Learning agent in PyTorch trained to play Connect 4, with experience replay, target networks and epsilon-greedy exploration. Result: DQN agent consistently beats random baseline (>70% win rate).
- **Personal Investment Portfolio — AI-Assisted** (2025 – Present) — Active management of personal investment portfolio with a custom-built dashboard (Streamlit + yfinance + Plotly + pandas): evolution analysis, market cap comparison, ARS/USD/crypto normalization. Built in intensive collaboration with AI — a natural extension of the AI-native practice from work into day-to-day financial decisions.
- **Facial Recognition with Raspberry Pi** (2023) — Implemented a facial recognition system (access control) on Raspberry Pi using OpenCV + TensorFlow Lite with embeddings.

## Skills

- **Languages:** Python (expert), Go, JavaScript / Node.js, Java (Spring Boot), C / C++, R, Assembler, SQL
- **AI & Data:** Agentic Workflows, RAG Architectures (Retrieval-Augmented Generation), Machine Learning (CatBoost, XGBoost, regression, clustering), Deep Learning (PyTorch, TensorFlow, neural networks), NLP (Hugging Face transformers, spaCy), scikit-learn, LLM Integration (Claude/Anthropic, OpenAI, Gemini, Whisper), AWS Textract (structured document extraction), Apache Spark (MapReduce, scalable pipelines), Monte Carlo simulation, Weights & Biases (experiment tracking)
- **Cloud & Infrastructure:** Linux (Pop!_OS daily driver; Ubuntu, Debian — advanced use; Kali Linux in UTDT's Cybersecurity course, CTF-style attack/defense labs), Google Cloud Platform (GCP — Cloud Run, Cloud Tasks, Functions), AWS (Textract for document extraction), Hetzner (VPS, self-hosted on-premise), Firebase / Supabase, Docker / Docker Compose (heavy production use), Kubernetes (academic training, UTDT infra courses), Raspberry Pi (on-premise server for thesis and real projects), Lambda / Serverless architecture, Cloudflare zero-trust tunnels, CI/CD (GitHub Actions, auto-deploy via SSH to VPS), Observability & Monitoring, Microservices
- **Backend & APIs:** Python (FastAPI, SQLAlchemy 2, Alembic — production migrations), Node.js / Express, Java Spring Boot, Firebase Functions, RESTful APIs (design & development), Legacy-system integration (SOAP/XML), n8n (workflow orchestration)
- **Frontend:** React, Svelte, Next.js, HTML / CSS, Vite
- **Databases:** PostgreSQL, MySQL, MongoDB, Firebase Firestore, Redis, Cassandra, Relational algebra
- **Product & Methodology:** 100% AI-native development, Agile / Scrum, Product Discovery, Technical consulting

## Education

- **BSc in Digital Technologies (Major in Computer Science, Minor in Business)** — Universidad Torcuato Di Tella (2022 – 2025). GPA: 9.13 / 10. Completed 39 courses (7 electives in AI, Infra, and Data Engineering — 7 beyond the 32 required to graduate)
- **Certified Tech Developer (Backend)** — Digital House (2021 – 2023). Full scholarship by Mercado Libre and Globant
- **Electromechanical Engineering (3 years completed)** — Universidad Nacional de San Juan (2018 – 2021). Transferred to Digital Technologies after discovering a passion for computing. Strong foundation in basic sciences and systems thinking retained from the engineering track.
- **Saint Paul — Bilingual English Program** — Saint Paul (San Juan, Argentina) (2012 – 2017). Bilingual foundation underpinning current C1 proficiency.

## Languages

- Spanish: Native
- English: C1 – Fluent
- German: A2 – Basic

## Achievements

- 2nd place in university Machine Learning competition (final model: CatBoost)
- Full scholarship from Mercado Libre and Globant for Certified Tech Developer (2021–2023)
- GPA 9.13/10 — BSc in Digital Technologies, UTDT
- Chosen by the CTO to evaluate and lead company-wide AI-tooling adoption — 100% of the company (~15 people) onboarded
- Teaching Assistant across 3 core Computer Science courses at UTDT since 2023