William Rodriguez (Wisrovi) | Principal AI Engineer & Suite Creator

No-Code Visual Builder Pro-Code Python Output Scratch / Blockly Blocks

WPipe Studio: No-Code Visual Pipeline Builder

The visual block-programming environment for the wpipe ecosystem. Design reactive data & AI pipelines by snapping visual blocks without writing code (100% No-Code), and compile with a single click into world-class production Python code (Pro-Code), FastAPI microservices, and automated tests. Zero vendor lock-in.

No-Code Block Programming: Assemble complex pipelines Scratch/Blockly-style. Connect triggers, AI inference (WYOLO), Kafka streaming, and persistence without syntax overhead.
World-Class Production Python Code: Instant bi-directional compilation. Generates typed, clean, production-grade WPipe code adhering to top global engineering standards—audit-ready and Git-native from day one.
Automated Microservice & Testing: Export your project as a full FastAPI REST service, Pydantic DTO validators, and complete pytest test suites in one step.
Lifetime Pro License ($49): Free tier with unlimited use (up to 5 blocks) and lifetime Pro license with unlimited blocks and real-time execution telemetry.

Get Lifetime Pro ($49) Explore PyPI Suite

wpipe-studio

Visual Block Canvas (No-Code) DAG Auto-Validation: Active

Scroll or drag to zoom & pan • Click ⛶ for fullscreen

Interactive Canvas: Zoom & Pan enabled (Click ⛶ for 1104x745) View generated Python code →

app/pipelines.py 100% Native WPipe

# Auto-generated from WPipe Studio No-Code Canvas
from wpipe import WPipe, PipeConfig
from wpipe_plugins.vision import ImageECamYOLO
from wkafka import WKafkaConsumer
from wsqlite import WSQLite

# Reactive pipeline without GIL bottleneck
pipeline = WPipe(
    name="surveillance_flow",
    config=PipeConfig(tracking=True)
)

pipeline.set_steps([
    WKafkaConsumer(topic="camera_feed"),
    ImageECamYOLO(model_path="yolo11n.pt", conf=0.45),
    WSQLite(db_path="audit.db", wal=True)
])

if __name__ == "__main__":
    result = pipeline.run({"device": "cuda:0"})

Compatible with pytest, Docker, and native CI/CD

WPipe Studio v2.5.10 Zero-Code Assembly • 100% Python Export

Verified Academic Record

Scientific Research & Doctoral Aspirations

My research agenda combines formal mathematical rigor with mission-critical software engineering. Holding a Master's degree in Artificial Intelligence (VIU, Spain) and a Bachelor's in Electronic Engineering (UDI, Colombia), I pursue doctoral research across three primary frontiers:

Causal Saliency Regularization (CSR): Analytical mitigation of spurious correlations via truncated Jacobian penalty across the multi-scale PAFPN pyramid.
Conformal Prediction & Uncertainty Quantification: Finite-sample distribution-free coverage guarantees coupled with variational uncertainty decomposition via MC Dropout (T=30).
Formal Verification of Agentic Protocols: Model checking and LTL (Linear Temporal Logic) monitor synthesis for mathematical safety validation of FastMCP servers.

View Publications on ORCID

Preprint Portfolio (CERN/Zenodo)

Scientific Preprints

100%

Registered DOIs

IEEE

Canonical Format

Preprints experimentally investigate MLOps scalability, FID latent domain shift quantification, multi-scale Grad-CAM forensic audits, and formal verification in heterogeneous clusters.

                        ORCID ID: 0009-0005-0710-1861
Affiliation: wisrovi-suit AI Research Initiative, Badajoz, Spain
                    

Code	Manuscript Title	Research Frontier	Official DOI & Details
RND-01	Predicting Downstream Performance Degradation in Computer Vision Models Under Representation-Level Domain Shift	Domain Shift & Latent Features	10.5281/zenodo.22716928
RND-02	An Autonomous Evaluation Framework for Quantitative Robustness, Faithful Explainability, and Failure-Mode Auditing in Vision Models	Explainable AI (XAI) & Saliency	10.5281/zenodo.22716935
RND-03	Agentic MLOps: LLM-Driven Orchestration of Distributed Vision Training Clusters via Model Context Protocol	Agentic MLOps & FastMCP	10.5281/zenodo.22715752
RND-04	Automated Dataset Profiling and Exploratory Validation in the wyoloservice2 Stack	Explainable AI (XAI) & Saliency	10.5281/zenodo.22715756
RND-05	Quantifying YOLO Model Robustness — Adversarial Attacks, Noise Resilience, and Uncertainty Estimation	Adversarial Robustness & Noise	10.5281/zenodo.22715764
RND-06	Automated Cross-Domain Generalization Assessment for YOLO Object Detection: FID-Based Domain Shift Quantification and Hardware Profiling	Domain Shift & Latent Features	10.5281/zenodo.22715775
RND-07	LLM-Powered Automated Research Reporting for Computer Vision Pipelines: From Training Metrics to Scientific Narratives	Agentic MLOps & FastMCP	10.5281/zenodo.22715785
RND-08	Decentralized MLOps Architecture for Autonomous Training and Integrated Explainability	Explainable AI (XAI) & Saliency	10.5281/zenodo.22715789
RND-09	Beyond Metrics: Autonomous Evaluation of Neural Networks via Grad-CAM and Local LLMs	Explainable AI (XAI) & Saliency	10.5281/zenodo.22715803
NP-01	NeuralForge: A Distributed MLOps Framework for Automated YOLO Hyperparameter Optimization	Distributed MLOps Architecture	10.5281/zenodo.22715816
NP-02	Industrial Experience Report: The Invoker-Executor Pattern for Fault Isolation in Distributed YOLO Training	Distributed MLOps Architecture	10.5281/zenodo.22715827
NP-03	WPipe: A Declarative Pipeline Framework with Forensic Tracking for ML Workflows	Distributed MLOps Architecture	10.5281/zenodo.22715839
NP-04	Outlier Failure Analysis: A Data-Centric Approach to Hard-Negative Mining in YOLO	Distributed MLOps Architecture	10.5281/zenodo.22715846
NP-05	Automated Hardware Complexity Profiling for Edge-Deployed YOLO Models	Distributed MLOps Architecture	10.5281/zenodo.22715852
NP-06	Automated Explainable AI Pipeline for YOLO Models: From Grad-CAM to Quantitative Fidelity Validation	Explainable AI (XAI) & Saliency	10.5281/zenodo.22715858
NP-07	Quantifying YOLO Model Robustness: Adversarial Attacks, Noise Resilience, and Uncertainty Estimation in Real-World Deployment	Adversarial Robustness & Noise	10.5281/zenodo.22715864
NP-08	Robustness and Noise Evaluation for Real-World Object Detection: Defending Against Sensor Degradation	Adversarial Robustness & Noise	10.5281/zenodo.22715870
NP-09	Statistical Rigor in YOLO Model Evaluation: Bootstrap Confidence Intervals and Failure Mode Analysis	Statistical Rigor & Bootstrapping	10.5281/zenodo.22715887
NP-10	Cross-Domain Generalization Assessment for Object Detection Models: FID-Based Domain Shift Detection	Domain Shift & Latent Features	10.5281/zenodo.22715899
NP-11	LLM-Assisted Training Report Generation in a Distributed MLOps Pipeline: Architecture, Deterministic Fallback, and Empirical Evaluation	Agentic MLOps & FastMCP	10.5281/zenodo.22715914
PHD-01	Self-Healing Distributed MLOps: Autonomous Fault Detection Node Eviction and Resilient State Reconciliation	Doctoral Frontiers & Formal Verification	10.5281/zenodo.22716656
PHD-02	Formal Verification of Agentic MLOps Protocols: Model Checking and Liveness Guarantees in Distributed Model Context Protocol Workflows	Doctoral Frontiers & Formal Verification	10.5281/zenodo.22716667
PHD-03	Causal Saliency Regularization: Mitigating Clever Hans Artifacts in Deep Object Detectors via In-Training Explainability Penalties	Doctoral Frontiers & Formal Verification	10.5281/zenodo.22716676
PHD-04	Conformal Prediction and Epistemic Uncertainty Decomposition for Safety-Critical Edge Vision under Dynamic Domain Shift	Doctoral Frontiers & Formal Verification	10.5281/zenodo.22716681
PHD-05	Carbon-Aware Edge Shrinkage: Dynamic Head Pruning and Adaptive Voltage-Frequency Scaling for Green Object Detection	Doctoral Frontiers & Formal Verification	10.5281/zenodo.22716692
PHD-06	Zero-Shot Diffusion Purification: Stochastic Denoising Defense for YOLO Detectors against Transferable Adversarial Attacks	Doctoral Frontiers & Formal Verification	10.5281/zenodo.22716695

Official Suite Catalog (34 Tools & Packages)

All 34 official suite packages, tools and FastMCP servers actively maintained and installable via the automated w-cli terminal suite.

Ctrl+K

PRO / FREE VISUAL IDE

WPipe Studio

Visual Pipeline Canvas & IDE Extension. Scratch/Blockly-style node graph editor with automated DAG compilation, live telemetry and native wpipe-engine execution.

VS CodeExtension

Free / ProEd25519

CORE SUN

wpipe

High-performance Python pipeline orchestrator (Sync/Async). Declarative DAG core engine with real GIL-bypass parallelism, SQLite WAL checkpoints, and contextual dependency injection.

CoreEngine

Sync/AsyncDual Mode

Streaming

wkafka

Decorator-based high-throughput wrapper for Apache Kafka. Non-blocking consumers/producers with automated reconnection, thread pools, and backpressure handling.

DecoratorsAPI

Auto-ReconnectBuilt-in

Cache & Locks

wredis

Ultra-fast Redis sync/async integration & caching suite. Connection pooling, token-bucket distributed rate limiters, pub/sub streams, and atomic distributed locks.

<1msLatency

PubSubDistributed

Resilience

wdecorators

Reusable high-level decorator suite. Implements exponential backoff with full jitter, runtime latency profiling, memory tracing, and structured Loguru logging.

Retry JitterResilience

Latency ProfilerBuilt-in

FastMCP

wpipe-mcp

Official FastMCP Server for wpipe pipeline architecture. Enables LLM agents (Claude, Cursor, Antigravity) to design, inspect, formally verify, and dry-run DAG execution flows.

FastMCPProtocol

AgenticTools

Relational

wpostgresql

High-concurrency PostgreSQL connection pooler and enterprise ORM mapper. Thread-safe connection pools, ACID transaction decorators, and Pydantic model synchronization.

ACIDPool Ready

ORMMapper

Modular Steps

wpipe-steps

Official catalog of production-ready modular steps for wpipe. Pre-built, strongly typed steps for data ingestion, ETL, computer vision, and ML transformations.

StepsLibrary

ModularArchitecture

Embedded DB

wsqlite

High-performance SQLite ORM wrapper featuring reactive TableSync. Optimized multi-thread Write-Ahead Logging (WAL) mode, connection pooling, and automated schema migration.

WAL ModeThread-Safe

TableSyncAuto

Computer Vision

wyolo

Complete lifecycle training wrapper for YOLO (v8, v11, YOLO26) and RT-DETR. Seamless synchronization of weights to MinIO S3 and real-time metric logging to MLflow.

YOLOMLOps

MLflow & S3Sync

FastMCP Cluster

wyoloservice-mcp

FastMCP Server for NeuralForgeAI YOLO cluster orchestration. Enables LLMs to validate datasets, generate Sweeper v2 configurations, inspect GPU queues, and dispatch Optuna sweeps.

FastMCPServer

ClusterOrchestrator

Credentials Vault

wauth

Cryptographic secrets encryption vault locked to host hardware fingerprints (machine-salted). AES-256 Fernet symmetric encryption protecting sensitive cluster configurations.

AES-256Fernet

Hardware-LockedVault

Zero Trust

wFabricSecurity

Industrial Zero Trust security architecture for Hyperledger Fabric networks. ECDSA P-256 digital signature validation, SHA-256 code integrity hashing, and cryptographic audits.

Zero TrustModel

HyperledgerFabric

Audio ML / DSP

ProcessAudio

Scikit-learn compatible transformers for audio feature extraction (Mel spectrograms, MFCC, FFT spectral centroid) and real-time acoustic data augmentation pipelines.

DSP/FFTFeatures

Scikit-learnPipes

OLAP Analytics

wclickhouse

High-throughput ClickHouse ORM mapping using clickhouse-connect. Optimized for bulk batch insertions, columnar time-series aggregation, and deep learning inference telemetry.

OLAPBatch Insert

Time-SeriesFast

Docker SDK

wcontainer

Docker SDK automation engine: ephemeral container spawner, GPU VRAM & CPU quota enforcement governors, and embedded Trivy vulnerability scanning.

Docker SDKAutomation

GPU IsolationGovernor

Document NoSQL

wmongo

Reactive MongoDB document mapper utilizing Pydantic models. Features automatic query caching on wredis, safe aggregation pipelines, and lifecycle event listeners.

PydanticModels

AggregationPipelines

FastMCP Kafka

wkafka-mcp

FastMCP Server for Kafka event streaming and topic inspection. Exposes tools for LLM agents to monitor topic partitions, consumer group lag, and inject telemetry payloads.

FastMCPServer

KafkaStream Ops

FastMCP Redis

wredis-mcp

FastMCP Server for Redis cache management and memory profiling. Enables AI agents to safely query keys, inspect data structures, and monitor memory fragmentation.

FastMCPServer

RedisOps

FastMCP Postgres

wpostgresql-mcp

FastMCP Server for PostgreSQL enterprise databases. Gives LLM agents safe, parameterized query tools, schema inspection capabilities, and role permission validation.

FastMCPServer

PostgreSQLOps

FastMCP SQLite

wsqlite-mcp

FastMCP Server for SQLite database inspection and queries. Parametric safe queries, index statistics analysis, and audit state extraction for wpipe pipelines.

FastMCPServer

SQLiteQueries

Plugins Framework

wpipe-plugins

Community-driven plugin repository and hot-reloadable extension framework for wpipe. Dynamic dependency isolation for custom enterprise pipelines.

PluginsFramework

ExtensibleArchitecture

Utilities & Cron

wutils

Essential script scheduling and validation helpers. Robust cron expression parser, dataset format verification, and algorithmic data normalization.

Task SchedulerCron

Format ValidatorFast

Core Foundational

wisrovi-python

Core foundational Python types, algorithms, and mathematical structures shared across all packages in the wisrovi ecosystem. Optimized for inference efficiency.

Core TypesShared

Math HelpersOptimized

NoSQL TinyDB

wtinydb

TinyDB Pydantic document mapper and lightweight query suite. Zero-config embedded document persistence for edge and microservice state.

EmbeddedStorage

Pydantic v2Schema

FastMCP Agent

wtinydb-mcp

Official FastMCP Server for WTinyDB. Provides autonomous LLM agents with structured document retrieval, table inspection, and state queries.

FastMCPProtocol

EmbeddedTools

Vector DB

wmilvus

High-throughput Milvus vector database client and similarity search suite. Distributed embeddings indexing, collection management, and cosine/L2 distance search.

VectorSearch

EmbeddingsScalable

FastMCP Agent

wmilvus-mcp

Official FastMCP Server for Milvus. Equips AI agents with vector collection querying, hybrid dense/sparse search, and semantic similarity tooling.

Vector MCPTools

RAG / SearchHybrid

MySQL ORM

wmysql

High-throughput MySQL Pydantic ORM wrapper. Connection pooling, declarative migrations, schema verification, and transactional consistency.

MySQLPooling

ACIDTransactions

MariaDB Suite

wmariadb

Enterprise MariaDB connection pooling and query builder. High-availability cluster routing, Pydantic model serialization, and thread-safe operations.

MariaDBCluster

HA PoolResilient

FastMCP Agent

wmariadb-mcp

Official FastMCP Server for MariaDB. Enables autonomous LLM agents to inspect database catalogs, query tables, and diagnose replication status.

FastMCPProtocol

SchemaAuditing

FastMCP Agent

wclickhouse-mcp

Official FastMCP Server for ClickHouse. Exposes columnar analytical queries, real-time log exploration, and aggregation benchmarks to AI agents.

OLAP MCPAnalytics

ColumnarReal-time

Cloud DW

wsnowflake

Dynamic Snowflake SQL client and schema mapper. Managed data warehouse queries, async cursor fetch, and Arrow/Pandas DataFrame extraction.

SnowflakeCloud

DataFramesFast Fetch

Messaging Suite

wmessenger (wconnect)

Enterprise messaging library with Telegram and Slack integration. Automated pipeline alerts, bot commands, and resilient webhook dispatchers.

TelegramBot API

SlackWebhooks

William Rodriguez

1. Production Engineering

2. Wisrovi Suite (PyPI)

3. Scientific Research

The Binary Universe of the Wisrovi Suite

Python Core Ecosystem & MLOps

FastMCP Hub for LLM Agents

WPipe Studio: No-Code Visual Pipeline Builder

wyoloservice: Distributed Training Platform

NeuralForgeAI Distributed Lifecycle Architecture

Scientific Research & Doctoral Aspirations

Preprint Portfolio (CERN/Zenodo)

Official Suite Catalog (34 Tools & Packages)

WPipe Studio

wpipe

wkafka

wredis

wdecorators

wpipe-mcp

wpostgresql

wpipe-steps

wsqlite

wyolo

wyoloservice-mcp

wauth

wFabricSecurity

ProcessAudio

wclickhouse

wcontainer

wmongo

wkafka-mcp

wredis-mcp

wpostgresql-mcp

wsqlite-mcp

wpipe-plugins

wutils

wisrovi-python

wtinydb

wtinydb-mcp

wmilvus

wmilvus-mcp

wmysql

wmariadb

wmariadb-mcp

wclickhouse-mcp

wsnowflake

wmessenger (wconnect)

Install the Suite with w-cli

Preprint Title