Portrait of Dennys Heras

A little about me

Software, automation, bots and AI agents for businesses.

More than 8 years in fiber optics
Remote · Spanish native · English B1

Write to me

Code

$ cat ruleset.v1.json
{ "id": "R-02",
  "name": "intentos-en-rafaga",
  "action": "Limit", "limitMinutes": 15,
  "condition": { "type": "burst",
    "min": 10, "maxWindowSeconds": 60 } }

$ audit static-example
rule        R-02 v1.0.0
decision    Limit
used        account, failed attempts,
            time window
not used    name, nationality, language,
            device type, gender, source IP
review      PendingReview

What I can do, the rules I follow when working with AI, and an audit example to see it in practice.

Projects

A Telegram bot that more than 10 technicians at an ISP use. A children's football championships site for a client. Criden, for businesses and websites.

Telegram botsSee the 4 bots
  1. Box registry botIn developmentRegisters network boxes from the field, with location and evidence, to build a more reliable technical inventory.
  2. Nearest box botIn productionDevelopment and launch (2025) of a Telegram bot for ISP technicians that finds the nearest network box to their location.
  3. Reports botIn developmentTechnical telecom reports from Telegram: incidents, inspections and follow-up.
  4. Planning botIn developmentReassigns activities in the team's Telegram group: it notices when someone asks to reschedule, proposes a free crew and time slot, and waits for the technician to confirm.

From fiber to AI

More than 8 years in fiber-optic networks, where things have to work in the field.

How I work

People

  1. Tests first, then code.
  2. Nothing is assumed without evidence: an output, a passing test.
  3. Adversarial review of security, tests and correctness before publishing.
  4. Real data only. What I do not have is marked as not yet confirmed; never invented figures or achievements.
  5. I authorize commits and publishing. No agent publishes on its own.
  6. Every decision is written down: date, alternatives and reason.
  7. No AI model call per visitor click.

Machines

# Working rules (llms.txt format)
- Tests first, then code.
- Nothing is assumed without evidence: an output, a passing test.
- Adversarial review of security, tests and correctness before publishing.
- Real data only. What I do not have is marked as not yet confirmed; never invented figures or achievements.
- I authorize commits and publishing. No agent publishes on its own.
- Every decision is written down: date, alternatives and reason.
- No AI model call per visitor click.

About me

Who I am, what I do and how to reach me.

Explore

  1. Code

    What I can do, the rules I follow when working with AI, and an audit example to see it in practice.

    Open Code → (opens a new page)
  2. Projects

    A Telegram bot that more than 10 technicians at an ISP use. A children's football championships site for a client. Criden, for businesses and websites.

    Open Projects → (opens a new page)
  3. From fiber to AI

    More than 8 years in fiber-optic networks, where things have to work in the field.

    Open Journey → (opens a new page)
  4. How I work

    I use AI to move faster, but nothing gets published on its own: I review and approve every change myself.

    Open How I work → (opens a new page)
  5. About me

    Who I am, what I do and how to reach me.

    Open About me → (opens a new page)