Ask about Deven

AI assistant with context about my work

Ask me anything about Deven's work, skills, or experience.

Try asking

Powered by AI with context about Deven's work

00.0043.65°NSYS.READY79.38°W
Online
|
|
LOC:TORONTO.CASTACK:TS/PY/REACT/NODEEDU:UWATERLOO.BMATH.2025FOCUS:AI.SYSTEMS.INFRASTATUS:SEEKING.ROLESBUILD:CONCURRENT.AIEXP:FULL-STACK.AI.PRODUCTCLEARANCE:OPENLOC:TORONTO.CASTACK:TS/PY/REACT/NODEEDU:UWATERLOO.BMATH.2025FOCUS:AI.SYSTEMS.INFRASTATUS:SEEKING.ROLESBUILD:CONCURRENT.AIEXP:FULL-STACK.AI.PRODUCTCLEARANCE:OPEN

Deven

Liscombe

Profile

Engineer building end-to-end AI systems across product, infrastructure, and agents.

Looking for an early-stage or AI-native team where I can own systems end-to-end — working closely with product and founders on ambiguous problems.

Readout

Focus

AI Systems

Infra + Product

Education

UWaterloo

BMath '25

Primary

TS / PY

Full Stack

Scope

End-to-End

Solo Builds

Active2024 – Present

Founder / AI Systems Engineer

  • 01Designed and built an end-to-end AI system for persistent context and memory in agent workflows — solving the core failure mode of modern AI: statelessness
  • 02Implemented agent runtime, context graph (structured memory), orchestration layer, and MCP server as sole engineer
  • 03Shipped consumer web app, desktop + mobile apps, and internal SDK across a 12-week build

Minto Group

2023

Financial Analyst

  • Oversaw finances and cash flow projections for commercial real estate portfolio
  • Built VBA-automated budget templates, replacing manual workflows

Enwave

2022

Corp Dev Finance Intern

  • Financial modelling for optimal pricing on historical electricity rate data
  • Rebuilt data archive into dashboard format for deal pricing
University of Waterloo

University of Waterloo

BMath (Honours Mathematics)

2021 – 2025

Majors in Mathematical Finance and Statistics.

Measure Theory
Stochastic Processes
Time Series
Numerical Methods

Deep Experience

TypeScriptPythonPostgreSQLNode.jsReactNext.jsAgentsMemory SystemsElectronMCP

Experience With

LLM APIsRAGpgvectorRedisDockerAWSReact NativeTailwindGoML Pipelines

Concurrent

A system for persistent intelligence — a cognitive substrate that lets AI reason, remember, and act coherently across time, tools, and tasks. Solving the core failure mode of modern AI: statelessness.

RoleSole Engineer
ScopeFull Stack + AI
Duration12 Weeks
StatusProduction
app.tryconcurrent.ai
ARCH.01
CONCURRENT.SYS
PRESENTATIONSERVICEINTELLIGENCEPERSISTENCESURFACEUser InterfacesMulti-platform product surfacesAPIBackend APIsREST APIs with modular servicesCOREAgent OrchestratorReAct reasoning with auditable stateMEMORYContext GraphStructured memory for persistent intelligenceTOOLSMCP ServerExternal tool and agent integrationSTOREData LayerPersistent storage and retrievalClick any node for details
Product SurfacesPROD
  • 01Consumer web app (primary)
  • 02Marketing website
  • 03Desktop + mobile apps
  • 04Operator Studio (SDK + UI)
Core InfrastructureINFRA
  • 01Context Graph memory system
  • 02Agent Orchestrator (ReAct)
  • 03MCP Server integration
  • 04Backend APIs (REST)
Data & AI SystemsDATA
  • 01PostgreSQL schemas
  • 02Vector embeddings (semantic)
  • 03Persistent session storage
  • 04Memory-scoped inference

Quantitative Trading System

Python · PostgreSQL · Event Streaming · ML Pipelines

PRIVATE REPO

Full-stack trading infrastructure with clean separation between research, execution, and risk systems. Live trading consumes the same feature definitions as research — just computed in real-time.

Data Layer

  • 01Raw → Normalized → Derived pipeline
  • 02Timeseries store for bars/ticks
  • 03Corporate actions + calendar
  • 04Versioned dataset snapshots

Research + Strategy

  • 01Event-driven backtest engine
  • 02Realistic cost + slippage models
  • 03Point-in-time ML training
  • 04Model registry + versioning

Execution + Risk

  • 01OMS with lifecycle state machine
  • 02Pre/post-trade risk checks
  • 03Position limits + drawdown stops
  • 04Broker reconciliation service

Open to engineering roles at early-stage and AI-native teams. Prefer to work on ambiguous, high-leverage problems end-to-end.

devenliscombe.dev
Next.js + Tailwind·Toronto, Canada·43.65°N 79.38°W