Gabriel Valerio · Business automation

Systems that work while you sleep.

I help teams replace repetitive manual work with systems that keep running after the workday ends. I connect the tools you already use, automate routine decisions, and make important information easier to act on. The activity feed shows examples of the work these systems handle.

[email protected] View work ↓

50+workflows in production

6industries served

24/7monitoring and alerts

EXAMPLES · SYSTEM ACTIVITY

ILLUSTRATIVE

EXAMPLES OF AUTOMATED WORK · CLIENT DETAILS OMITTED

Work that changed the workflow

10 PROJECTS

Real production systems, described by the problem they solved and what changed afterward. Company details are anonymized where needed; numbers are real.

CLIENT: SPORTS-TECH SAAS (US) 2026 — NOW IN PRODUCTION

Gave an AI agent reliable access to company knowledge

A sports-tech company had important knowledge spread across Salesforce, playbooks, past cases, and meeting transcripts. I built a private search system that brings the right context together before an AI agent answers, instead of making the team search across disconnected sources. It runs on a local GPU, so data stays on the machine and queries do not create a per-use API bill.

STACK: PYTHON · SQLITE-VEC · FTS5 / BM25 · QWEN3 EMBEDDINGS · ENTITY GRAPH

1,018knowledge units indexed

1,477knowledge connections mapped

12s → 0.13ssearch response time, cached

CLIENT: SPORTS-TECH SAAS (US) 2026 — NOW IN PRODUCTION

Handled routine support questions without sacrificing trust

A sports-tech support team was receiving repetitive questions through chat and email. I built an AI helpdesk agent that answers only from approved company information and sends billing, legal, or uncertain questions to a person. The team can absorb routine volume without hiring, while FERPA-scoped data remains on US-pinned model hosting.

STACK: NODE / EXPRESS · SUPABASE · RAG · OPENROUTER · HMAC WEBHOOKS

99%correct on 162-case eval

0hallucinations · 100% grounded

2channels · chat + email

CLIENT: ENTERPRISE RETAIL 2026 DELIVERED

Kept two project systems in sync automatically

A retail team was maintaining the same work in Monday.com and Azure DevOps. Updates had to be copied by hand, creating delays and inconsistencies. I built a two-way connection so Epics and Features move between both systems automatically, without duplicate updates or sync loops.

STACK: MAKE.COM · MONDAY GRAPHQL v2 · AZURE DEVOPS API

~90%manual sync ops cut / sprint

<5swebhook-to-write latency

7scenarios · bidirectional

CLIENT: COMMODITIES (LATAM) 2026 DELIVERED

Cut unnecessary automation usage without interrupting operations

A commodities company’s automation bill had grown because one workflow triggered too often and repeatedly retried failed runs. I found the cause, rebuilt 25 automations using the platform’s native features, and checked the new workflows against the old ones before switching over. The migration happened with zero downtime.

STACK: MAKE.COM · MONDAY.COM NATIVE · NODE / EXPRESS POC

39,679/moautomation operations recovered

25automations rebuilt with native features

0pipeline disruptions

CLIENT: INVESTMENT MANAGEMENT 2026 DELIVERED

Made compliance checks easier to review

An investment manager needed a repeatable way to screen counterparties and keep the evidence organized. I connected the screening services, assembled structured PDF dossiers, and added them to Monday.com alongside automations for HR admissions and staff movements. The deployment process caught silent production bugs before they affected the workflow.

STACK: MAKE.COM · LSEG WORLD-CHECK ONE · NETRIN · MONDAY · DOCUSIGN

3silent go-live bugs caught

0false positives after fix

10+automations delivered

CLIENT: EDTECH (US) 2025 DELIVERED

Moved 14,000+ work items to a new system safely

An EdTech company needed to leave Monday.com without losing records or stopping the team’s work. I built a migration process that moved the data in controlled waves, checked every batch, and supported rollback if something went wrong. The team kept working while the system changed underneath them.

STACK: PYTHON · MONDAY API · JIRA CLOUD · ETL

14,000+items migrated

0records lost

3 wk → 4 dvs. manual plan

FLAGSHIP BUILD 2026 — NOW IN PRODUCTION

Turned daily work into a searchable second brain

This always-on system captures work and meetings as searchable text, finds action items, and drafts possible responses. It keeps only the text, deletes captured images immediately, and requires human approval before anything is sent or changed externally. The result is a private memory of work that can be searched instead of reconstructed from memory.

STACK: PYTHON · SQLITE-VEC / FTS5 · WHISPERX · OLLAMA · 22 SYSTEMD SERVICES

~800×storage reduction vs raw

61% → 100%retrieval hit@10, tuned

22services · self-healing

DATA PLATFORM 2026 — NOW DEPLOYED · MAINTAINED

Made election data useful at neighborhood level

Campaign teams needed more than a statewide view of election results. I built a platform that moves from state-level numbers down to individual voting locations, with maps for comparing areas and a console for managing field teams. It combines Brazil’s open government datasets and matches voting sections to neighborhood boundaries.

STACK: REACT 19 · DECK.GL / MAPBOX · FASTAPI · GOOGLE BIGQUERY · DOCKER / VPS

31municipalities modeled

4drill levels · state → location

8analytical views

CLIENT: CONTENT CREATOR 2025 — NOW IN PRODUCTION

Turned long videos into a weekly publishing pipeline

A content creator was spending around 15 hours each week finding clips, editing them, and writing captions. The system transcribes each video, identifies promising moments, cuts the clips, and prepares captions for TikTok, Reels, and Shorts without requiring manual editing for every post.

STACK: CLAUDE CODE · WHISPER · FFMPEG · PYTHON

36/wkclips produced

~15 h/wkediting eliminated

PERSONAL SAFETY SYSTEM 2024 — NOW IN PRODUCTION

Added an always-on safety alert around a pool

A home pool needed monitoring even when nobody was actively watching it. I connected camera detection to Telegram, email, and an on-site audio alert so the system warns people as soon as someone approaches the water. It has run autonomously since deployment.

STACK: PYTHON · TUYA CLOUD API · TELEGRAM BOT · AI DETECTION

<2salert latency

24/7autonomous uptime

Capabilities

04 AREAS

Code when it needs code, no-code when it doesn't. The right tool is the one your team can live with.

Workflow automation

I replace copying, chasing, and repetitive updates with workflows that move information between the tools your team already uses.

Data & analytics

I turn scattered or difficult-to-read data into dashboards, reports, and views that help teams decide what to do next.

AI integration

I connect AI to real work, with clear boundaries for what it can answer, change, or send and when a person must review it.

No-code systems

I build practical Airtable and Monday.com systems that your team can understand, operate, and extend after handoff.

What is your team still doing by hand?

Tell me where work gets copied, delayed, or repeated. I’ll help you identify what a reliable system could handle instead.

Send email LinkedIn