Skip to content

Latest commit

 

History

4 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 

Repository files navigation

Malcolm Frank

AI Infrastructure · Agentic Software Engineering · MLOps · Robotics

I build intelligent software systems that connect AI models, agents, data, infrastructure, automation, and real-world machines.

My engineering background spans software development, AI-native applications, robotic systems, cybersecurity, and distributed systems. My current focus is designing reliable agentic software and AI infrastructure that can plan work, use tools, maintain memory, retrieve knowledge, verify outcomes, and operate through observable, testable workflows.

Currently interested in opportunities across:

  • AI Infrastructure Engineering
  • Agentic Software Engineering
  • AI / ML Systems Engineering
  • MLOps & AI Platform Engineering
  • Robotics & Autonomous Systems
  • Systems Integration Engineering
  • Applied AI Engineering
  • AI Solutions Architecture

Location: United States
Portfolio: malcolm.expert
LinkedIn: linkedin.com/in/jamal-jones-70ba46425
GitHub: github.com/jamalfrnk
Email: malcolmfrank91@gmail.com


Current Engineering Focus

AI Infrastructure

I am focused on engineering the infrastructure around AI systems rather than treating the model itself as the entire product.

Areas of focus include:

  • LLM application architecture
  • Model and tool orchestration
  • Structured outputs and typed interfaces
  • Retrieval-Augmented Generation
  • Embeddings and semantic retrieval
  • Persistent AI memory
  • Model routing
  • Human-in-the-loop systems
  • AI evaluation and verification
  • Observability and tracing
  • AI application security
  • Local-first and cloud AI infrastructure
  • Cost-aware model utilization
  • Reliable long-running workflows

Agentic Software Engineering

I am particularly interested in software systems where AI can move beyond answering questions and participate in structured execution.

I work with concepts including:

  • Goal decomposition
  • Task graphs
  • Agent orchestration
  • Tool calling
  • Durable workflows
  • Specialized agent harnesses
  • Verification loops
  • Memory systems
  • Approval gates
  • Automation
  • Retry and recovery strategies
  • Human intervention
  • Evaluation-driven improvement
  • Persistent project state
  • MCP and tool integration patterns

My philosophy is that useful agentic systems should be observable, resumable, measurable, and verifiable, not simply autonomous.


J.A.R.V.I.S. — Agentic Second Brain

One of my primary ongoing projects is J.A.R.V.I.S., a second-brain and agentic operating system designed to create a durable intelligence layer across projects, files, research, software repositories, and computer-based work.

The system is being designed around the execution loop:

GOAL ↓ TASK GRAPH ↓ EXECUTION ↓ VERIFICATION ↓ MEMORY UPDATE ↓ VISIBILITY ↓ LEARNING ↺

Architectural areas being explored and implemented

  • Persistent semantic, episodic, and procedural memory
  • File-first project state
  • Agent and tool orchestration
  • Task graphs and dependency management
  • Retrieval and contextual knowledge assembly
  • AI evaluation infrastructure
  • Human approval and governance
  • Model routing
  • Execution verification
  • Observability and traceability
  • Automation and recurring workflows
  • Local-first execution
  • Dockerized and isolated environments
  • Security controls
  • Runtime and model portability
  • Continuous improvement from failures and successful workflows

The long-term objective is a system capable of helping manage complex technical projects while preserving enough context and evidence for another compatible AI system or human engineer to understand exactly what happened and continue the work.


Robotics & Edge Systems

My AI work is informed by hands-on experience with physical systems and deployed robotics.

My robotics and field-engineering background includes work with systems involving:

  • Linux
  • Raspberry Pi
  • ROS
  • Autonomous mobility systems
  • Cameras and computer-vision hardware
  • LiDAR
  • Sensors
  • Edge computing
  • Wi-Fi networking
  • DHCP and DNS
  • VPN and mesh networking
  • Tailscale / WireGuard
  • Remote diagnostics
  • Device provisioning
  • Fleet operations
  • Hardware/software integration
  • Telemetry
  • Field deployments
  • Failure analysis
  • Acceptance testing
  • Operational troubleshooting

This experience influences how I approach AI engineering: software eventually has to survive contact with real infrastructure, networks, hardware, users, and failure conditions.


MLOps & AI Platform Engineering

I am building deeper expertise around the systems required to move AI applications from experiments into reliable software.

Areas include:

  • Docker
  • Linux
  • Git and GitHub
  • CI/CD
  • Automated testing
  • Reproducible development environments
  • API-based model services
  • Evaluation pipelines
  • Environment isolation
  • Model/application observability
  • Deployment workflows
  • Infrastructure automation
  • Quality gates
  • Secrets management
  • Security controls
  • Failure recovery
  • Production-readiness testing

I am especially interested in the intersection between MLOps, DevOps, and agentic systems, where AI workflows must be treated as production software rather than demos.


Selected Projects

J.A.R.V.I.S.

Agentic Second Brain / AI Operating System

A long-horizon engineering project exploring durable memory, task orchestration, retrieval, tool execution, evaluation, observability, and self-improving workflows for AI agents.

Focus: Agentic AI · AI Infrastructure · Memory · RAG · Evals · Tool Use · Automation · Systems Architecture


Solana Vault Prototype

A security-focused Solana vault architecture built with Rust and Anchor, supported by a TypeScript SDK and Next.js application.

Engineering concepts demonstrated

  • PDA-controlled SPL custody
  • CPI authorization
  • Deposit and withdrawal accounting
  • Governance controls
  • Two-step authority rotation
  • Mint exposure limits
  • Emergency controls
  • Account versioning and migration
  • Negative-path and adversarial testing
  • Automated CI quality gates
  • Reproducible development environments
  • Architecture and operational documentation

Verification

  • 97 Rust tests
  • 117 TypeScript SDK tests
  • 122 dApp tests
  • Automated formatting, build, Clippy, testing, and audit checks

The system is explicitly documented as an educational, pre-audit prototype rather than an audited or production custody product.

Review the repository →


LiquidAlpha

A software and research platform exploring market infrastructure, trading systems, blockchain markets, risk analysis, and real-time financial data.

The project has included work around:

  • Market-data integration
  • Trading interfaces
  • Signal systems
  • Financial analytics
  • Web application architecture
  • Automated testing
  • Wallet integration
  • Blockchain market infrastructure

Review LiquidAlpha →


YieldSensei

A DeFi intelligence project combining financial market data, technical analysis, machine-learning concepts, Solana infrastructure, and automated information delivery.

Areas explored include:

  • Data pipelines
  • Market analytics
  • ML-assisted signals
  • Blockchain integration
  • Telegram delivery
  • DeFi risk analysis
  • Automated workflows

Robotics Operations Systems

My robotics work has included the integration and support of deployed systems involving embedded computers, wireless networking, cameras, sensors, mobility platforms, and remote fleet infrastructure.

I have also developed software aimed at improving field operations, visit tracking, operational records, and deployment workflows for robotic systems.

Focus: Robotics · Edge Computing · Linux · Networking · Field Engineering · Systems Integration


Core Engineering Stack

AI / Agentic Systems

LLMs · RAG · Agentic Workflows · Tool Calling · AI Memory · Model Routing · Structured Outputs · Evals · Automation · MCP Concepts

Machine Learning / Applied AI

Python · ML Systems Integration · Embeddings · Semantic Retrieval · Multimodal AI · Document Intelligence · AI Evaluation

Infrastructure / MLOps

Docker · Linux · CI/CD · GitHub Actions · Testing · Environment Management · Observability · Quality Gates

Software Engineering

Python · TypeScript · JavaScript · Rust · React · Next.js · Node.js · REST APIs · WebSockets

Data

PostgreSQL · Drizzle ORM · JSON · Structured Data · Data Pipelines

Robotics / Edge

Linux · Raspberry Pi · ROS · LiDAR · Cameras · Sensors · Wi-Fi · Tailscale · WireGuard

Security

Threat Modeling · Authentication · Authorization · Secrets Management · Negative-Path Testing · Network Security · Secure Architecture

Distributed Systems / Web3

Solana · Rust · Anchor · SPL · PDAs · CPI · DeFi · Wallet Integration · Smart Contract Security


Engineering Philosophy

I am most interested in systems that connect several layers of engineering:

DATA ↓ MODELS ↓ AGENTS ↓ TOOLS ↓ SOFTWARE ↓ INFRASTRUCTURE ↓ PHYSICAL SYSTEMS ↓ OBSERVABILITY ↓ FEEDBACK

I believe dependable AI systems require more than strong models.

They require:

clear architecture + structured state + reliable tools + verification + security + observability + evaluation + human control

That is the engineering space I am actively building toward.


Background

  • Bachelor of Science in Software Engineering
  • Master's studies in Cybersecurity
  • Graduate-level AI Systems Management studies
  • Certified Blockchain Architect
  • Certified NFT Developer
  • Certified Cryptocurrency Trader
  • Software engineering experience
  • Robotics and field systems engineering experience
  • Web3 / DeFi engineering experience
  • AI application and automation development
  • Technical education and curriculum development

Let's Connect

I am interested in working with teams building:

  • AI infrastructure
  • Agent platforms
  • AI-native developer tools
  • MLOps platforms
  • Applied machine learning systems
  • Intelligent automation
  • Robotics and autonomous systems
  • Edge AI
  • AI security
  • Developer infrastructure
  • Distributed systems

Links

Portfolio · LinkedIn · GitHub · Email

Recruiters, engineering leaders, and collaborators: malcolmfrank91@gmail.com

About

Config files for my GitHub profile.

Topics

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Used by

Contributors