Joshua Albert / Software Engineer
PDF Discuss a role

Denver, Colorado

Joshua Albert

Senior Software Engineer | AI Systems

Python · LangGraph · FastAPI · Neo4j · TypeScript · React

Senior software engineer with about 20 years of professional and independent development experience across data-intensive products, graph interfaces, enterprise integration, developer tools, and tested full-stack systems. Current work includes typed, replayable agent-evaluation workflows and, separately, provenance-aware evidence and retrieval tooling, using agents and graphs only where warranted and keeping system boundaries explicit enough to inspect, evaluate, and govern.

  • Visible Network Labs · source-backed production claim~10 minutes → ~10 seconds
  • Visible Network Labs · source-backed workInteractive graph and visualization interfaces
  • Independent capability · repository-inspectedReplayable agent evaluation with per-node traces
Discuss a role Download Résumé Explore selected work
Emailjoshua.ablert@socratesone.com LinkedInlinkedin.com/in/joshua-j-albert GitHubgithub.com/socratesone Main websitesocratesone.com

Selected engineering evidence

Concrete systems. Evidence boundaries stated.

Source-backed employer claims and repository-inspected independent work are labeled separately. Public corroborating artifacts remain a follow-up item.

Visible Network Labs · source-backed production claim · not publicly corroborated

~10 minutes to ~10 seconds

Data pipeline performance

Reduced processing time for a production CSV import by removing N+1 query patterns, batching database operations, memoizing repeated lookups, and hardening edge-case parsing.

Graph Interfaces

Visible Network Labs · source-backed production claim · not publicly corroborated

Up to ~5K nodes / ~20K edges

Interactive graph systems

Built Vue, TypeScript, D3.js, and Apache ECharts interfaces for relational datasets, with dynamic filtering, styling, grouping, and performance work for large network graphs.

Evaluation & Reliability

Independent capability demonstration · repository-inspected · public run artifact pending

Replayable agent evaluation

Agent workflow evaluation

Built a nine-node typed LangGraph pipeline with structured Pydantic validation, weighted model-assisted scoring, policy filtering, versioned prompts, and replay fixtures that produce per-node audit traces plus reliability, cost, and latency metrics per run.

Professional experience

Production product work and current independent engineering.

SocratesOne Development LLC

Independent Software Engineer - AI Systems

Denver, Colorado

- Present

Independent engineering · repository-inspected current work

Current work spans separate systems: EvidEngine and VimCommander, a typed LangGraph evaluation workflow, and FastAPI-backed internal applications.

  • Implemented typed LangGraph workflows that separate state, model calls, scoring, policy checks, selection, and replay evaluation, with per-node traces and run metrics.
  • Built FastAPI-backed internal applications with validated contracts, persistent data models, audit trails, and automated test coverage.
  • Developed provenance-aware evidence and retrieval tooling with deterministic artifacts, stable identifiers, exact/BM25 lookup, graph traversal, optional Neo4j integration, and MCP tools.

Product Engineering

Visible Network Labs

Software Engineer, Full Stack

Denver, Colorado

-

Prior employment · source-backed production claims · public artifacts pending

Full-stack product engineering for data-heavy network-analysis applications.

  • Reduced a production CSV import from roughly 10 minutes to roughly 10 seconds by removing N+1 query patterns, batching database operations, memoizing repeated lookups, and hardening edge-case parsing.
  • Built Vue, TypeScript, D3.js, and Apache ECharts interfaces for relational datasets, including dynamic filtering, styling, grouping, and performance work for large network graphs.
  • Implemented SAML 2.0 single sign-on and granular, role-scoped authorization for complex data-access roles and permissions.
  • Built and maintained unit and end-to-end test infrastructure with GitHub Actions pre-merge checks for data-heavy application workflows.

Flagship projects

Current work in evidence systems and developer tools.

These repository-inspected projects demonstrate evidence tooling and deterministic full-stack systems. Public implementation artifacts remain a follow-up item.

Agent Systems

Independent project · repository-inspected · early-stage v0.1.0

EvidEngine

Local-first evidence-graph tooling that turns documents, configuration, source anchors, and Git history into source-traceable artifacts, giving agent retrieval an auditing, visibility, and governance layer so teams can see and trust what agents pull in.

View EvidEngine overview
  • Implemented deterministic ingestion, extraction, linking, indexing, and query stages with stable identifiers and explicit run artifacts.
  • Modeled documents, segments, passages, entities, relationships, files, commits, and authors with provenance and honest source locators.
  • Added BM25 keyword search, exact-key lookup, graph traversal, optional Neo4j export and synchronization, and Model Context Protocol (MCP) tools.
  • Designed source-code ingestion as addressable anchors — usable as provenance for code review and software-delivery (SDLC) automation — rather than unsupported behavioral claims about what code does.

Status: early-stage, source-install, CLI-first software with no hosted service or local UI; semantic retrieval is experimental and off by default.

Developer Tools

Independent project · repository-inspected · active development

VimCommander

A browser space shooter built with React, TypeScript, and a full automated test suite, designed to teach Vim commands through keyboard-driven gameplay.

Open VimCommander live build
  • Built the game with TypeScript, React, Phaser, Three.js, Zustand, Zod, Vite, and Vitest.
  • Separated keyboard semantics from game behavior through an engine-independent mapping layer that routes high-level Vim actions.
  • Modeled editor modes and transitions with a pure, deterministic finite-state machine.
  • Used JSON Schema-defined bindings and shared game-client contracts to support alternate clients and control systems.
  • Added automated unit and integration tests for mappings, state transitions, user interface behavior, and game systems.

Status: active development; the tutorial covers a deliberate subset of Vim commands.

Core technical capabilities

Full-stack engineering for AI products and platforms.

AI systems

Python, LangGraph, LangChain Core, OpenAI API, tool calling, structured outputs, RAG, graph-backed retrieval, agent evaluation, run metrics, policy controls, workflow orchestration, context engineering, and MCP integration. Demonstrated in the evaluation workflow and EvidEngine.

Frontend and visualization

React, TypeScript, Vue, JavaScript, D3.js, Apache ECharts, Phaser, Three.js, responsive interfaces, network visualization, and GIS interfaces. Demonstrated in graph interfaces and VimCommander.

Quality and delivery

System design, performance analysis, unit, integration, and end-to-end testing, replay and regression evaluation, observability, GitHub Actions, CI/CD, Docker, and Google Cloud Platform. Demonstrated across current and prior work.

Earlier experience

Foundation in web platforms and integration.

CommonPlaces

Drupal Developer

Manchester, New Hampshire

-

  • Developed core components of an internal document-management intranet, including LDAP authentication, multi-tier permissions, content workflows, and Drupal application functionality.
  • Contributed to Drupal migration and commerce work involving custom content-management requirements and legacy integration.

Independent and Freelance Development

Full-Stack Developer

- , intermittent

  • Built and maintained custom web applications and content-management implementations for small-business and consumer clients using PHP, MySQL, Drupal, WordPress, and CodeIgniter.
  • Modernized legacy applications, implemented custom CRUD workflows, integrated third-party services, and introduced source control and deployment automation.

Contact and résumé download

Discuss an AI systems or full-stack engineering role.

For roles spanning AI products and platforms, agent evaluation, developer tooling, full-stack product engineering, data-intensive systems, and technical integration, contact Joshua directly.

Email Joshua Download Résumé
Email
joshua.ablert@socratesone.com
LinkedIn
Joshua Albert on LinkedInlinkedin.com/in/joshua-j-albert
GitHub
SocratesOne on GitHubgithub.com/socratesone
Website
socratesone.com
Top