Skip to content
View JohnnyWilson16's full-sized avatar
  • 22:41 (UTC -12:00)

Block or report JohnnyWilson16

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
JohnnyWilson16/README.md

Hi, I'm Johnny Wilson Dougherty πŸ‘‹

Typing SVG

Building reliable AI agents, LLM-powered data pipelines, and robust data-quality infrastructure.

LinkedIn Email GitHub


πŸ“Œ About Me

  • πŸ”­ I build Agentic RAG systems, self-healing data pipelines, and production data-quality infrastructure.
  • 🌐 100+ Merged PRs across open-source repositories including FreshData, Semantica (12k+ ⭐), and Tel-Agent.
  • ⚑ Core Contributor (90+ merged PRs) to FreshData β€” contributed semantic validation, out-of-core streaming, and automated PyPI CI/CD.
  • πŸ› οΈ Tech stack: Python, LangChain, CrewAI, ChromaDB, FastAPI, Pandas, Polars, DuckDB, Pytest.
  • 🎯 Available for Full-Time Remote/Onsite AI/ML Roles and Freelance GenAI/Data Engineering Projects.

πŸš€ Featured Projects

  • Autonomous financial report analysis moving past rigid top-k retrieval.
  • Agent dynamically formulates queries, evaluates evidence, and synthesizes grounded insights.
  • Multi-provider model layer (Groq, OpenAI, Gemini, Ollama) with 100% test coverage.

  • Detects upstream schema drift and heals column headers at runtime without crashing ETL jobs.
  • Layered resolution: 0ms exact match β†’ local signature cache β†’ heuristics β†’ LLM reasoning.
  • Zero-PII transmission protocol with automated dtype coercion and audit telemetry.

  • High-throughput data quality engine delivering 118,000 rows/sec corruption cleaning.
  • Implemented field-aware text sanitizers, privacy filters, and PyPI release workflows.

  • End-to-end customer complaint triage pipeline using supervised NLP classifiers.
  • Automated routing, sentiment breakdown, and interactive analytical reporting.

🌐 Open-Source Contributions

FreshData Semantica Tel-Agent

Project & Ecosystem Domain Accepted / Merged Contribution & Engineering Impact
FreshData
FreshCode-Org/freshdata
Data Quality & Out-of-Core Execution Core Contributor (90+ Merged PRs): Engineered semantic validation, context-aware per-cell field validation, and field-aware text sanitizers reaching 118,000 rows/sec at 1.00 F1 score. Implemented streaming data workflows, benchmark suites, and automated PyPI Trusted Publishing CI/CD (View 90+ Merged PRs).
Semantica
semantica-agi/semantica
⭐ 12.8k
Model Context Protocol (MCP) & Context Infra Nested metadata filtering (PR #1603 β€” Merged): Resolved nested metadata filtering and normalization in handle_query_decisions (#1247) for MCP-based semantic decision routing and knowledge retrieval.
Tel-Agent
Dpro-at/Tel-Agent
⭐ 330+
Conversational AI & Agent Channels Mattermost transport & i18n (PR #243, PR #244, PR #248 β€” Merged): Implemented the Mattermost channel transport for the open-source AI phone assistant agent, resolved gateway auth challenges, and added full internationalization support across committed locales.

πŸ› οΈ Technical Skills

Core & APIs

Python FastAPI SQL JSON

AI, LLMs & Vector Systems

LangChain CrewAI ChromaDB HuggingFace Ollama

Data Engineering & Analytics

Pandas NumPy Polars DuckDB

MLOps, Quality & Tooling

Pytest GitHub Actions Git Ruff


πŸ“ˆ GitHub Analytics


πŸ“¬ Let's Connect

⭐️ Built by Johnny Wilson Dougherty

Pinned Loading

  1. agentic-rag-system agentic-rag-system Public

    Production-ready End-to-End Agentic RAG System using LangChain, ChromaDB, and CrewAI

    Python

  2. indian-pharmacy-analytics indian-pharmacy-analytics Public

    SQL-first analytics on 253,973 Indian medicines β€” stock monitoring, expiry risk, seasonal demand, and supplier optimization. SQLite + pandas + Faker.

    Python

  3. self-healing-data-pipeline self-healing-data-pipeline Public

    Autonomous self-healing data pipeline framework using schema caching, heuristics, and LLM semantic reasoning

    Python

  4. FreshCode-Org/freshdata FreshCode-Org/freshdata Public

    The explainable, semantic data-cleaning layer for pandas & Polars β€” automated repairs, local AI intelligence, and full audit trails for fresh data

    Python 5 14