Jess
Mathematician & Software Engineer

I build systems that cannot afford to fail.

I'm a mathematician who ships. I build identity infrastructure, distributed systems, and low-level tooling across Go, Rust, Kotlin, and Python, plus the ML and research work that comes from the same math. I deliver commissioned systems including SAML/OIDC federation gateways, event-driven payment ledgers, sub-100ns feature-flag engines, and hybrid RAG retrieval. Several are running live below.

After I was honorably discharged from the Army, I started college on the Army's dime and got into discrete mathematics, which still shapes how I engineer. These days I'm open to full-time roles and contract work.

Get in touch View Projects LinkedIn GitHub
Selected Work

Featured systems

Titles link to the live, running application — click through to explore it firsthand.

01

RheoLogos live

A double-entry payment ledger with an event-driven outbox, an idempotent consumer, and a dead-letter queue. Kafka and Redis run alongside the app, Postgres is managed.

KotlinSpring BootKafkaPostgres
02

FlagForge live

A real-time feature flag and configuration engine. Single binary, embedded UI, written in Go (I also have a Rust version) nothing else to run.

Go
03

B2B Federation Gateway live

An enterprise identity gateway speaking SAML, OIDC, SCIM, and ABAC to each other, against a working demo tenant.

GoSAMLOIDCSCIM
04

NovaStructur live

A project manager that also tracks issues, not the other way around. Built to stay low bloat instead of turning into another Jira. It features a custom hand-rolled CRDT, an MCP for LLM integration, a tiny custom RAG with bespoke binary semantic search, and custom chunking logic specifically optimized for coding.

Next.jsPostgresCRDT sync
More projects →
Research & Development

Where the math gets tested

01

Physics Informed Neural Networks, Inverse Problems

Repeated a published PINN parameter recovery on Fisher-KPP, a reaction diffusion equation with a traveling front. Across 12 seeds the PINN missed the diffusion rate by 33% rms while a plain classical fit landed around 10%. Traced it to the loss itself scoring the wrong answer better than the right one, then salvaged the bad runs to about 16% by reading the parameter off the front's global behavior instead.

PyTorchPINNsInverse ProblemsFinite Differences
02

Graph Neural Network PDE Solvers

Trained GNNs to imitate classical solvers across elliptic, parabolic, and hyperbolic equations, each validated against a finite difference or spectral reference. Fixed a rollout instability on Burgers' equation with curriculum based unrolled training, and cut long horizon error by more than half with a Cole-Hopf transformation.

PyTorch GeometricGNNsPDEsCFD
03

Silicon, Tiny Tapeout

Took a design all the way down to silicon, completing a chip for the Jane Street competition on Tiny Tapeout.

ASICTiny Tapeout
Experience

Where I've shipped

Templet Solutions

2025 — Present
Software Engineer · commissioned client work and independent product builds
  • Built Limes Cross-Cloud Network Guard (Go) for multi-cloud CIDR diagnostics
  • Engineered B2B Enterprise Identity Gateway with hand-rolled XML-DSig & RS256 verification
  • Designed sub-100ns hot-path Feature Flag Engine with lock-free evaluation
  • Built hybrid RAG/GraphRAG retrieval engine with adaptive per-query weighting
  • Authored cryptographic audit & provenance system with Merkle-tree verification

TenantTransparency

2026 — Present
Freelance Software Engineer
  • Rebuilt tenant-rights platform from template to production in under one month
  • Shipped containerized React/Vite frontend with interactive neighborhood map
  • Built landlord entity-resolution pipeline

High Tech Systems and Software

2020 — 2025
Software Engineer
  • Engineered custom APIs and workflow automation for DataKlas CRM/SFA platform
  • Developed BI layer with layered ML models for consumer demand forecasting

Prior Experience & Military Service

2001 — 2020
Legal Assistant & U.S. Army Active Duty
  • Legal Assistant at Hartness Law Office; Dodson, Hooks & Frederick (Trial prep & Deepwater Horizon litigation support)
  • United States Army Active Duty — Honorably Discharged
Capabilities

Technical range

Languages & Tooling

  • Go, Rust, Python, Ruby on Rails 8
  • TypeScript, JavaScript, React
  • Axum, FastAPI, Tauri, Vite
  • tokio, Docker

Identity & Security

  • SAML 2.0, OIDC, SCIM 2.0
  • OAuth 2.1, PKCE, JWT/JWKS
  • ABAC/RBAC engines, XML-DSig
  • Cryptographic audit trails

Data & Hybrid Cloud

  • PostgreSQL (RLS), Kafka, Redis
  • AWS, GCP, Azure
  • BoltDB, Linux systems engineering
  • Distributed data infrastructure

Applied Mathematics

  • Hyperbolic geometry, spectral graph theory
  • Topological data analysis
  • Differential geometry
  • Discrete mathematics

Retrieval & Search

  • RAG, GraphRAG, hybrid retrieval
  • Dense embeddings + BM25 + graph traversal
  • ANN vector search (hnswlib)
  • sentence-transformers

Open Source Systems

  • binsg — semantic search CLI
  • turbogrep — zero-copy mmap grep
  • kvstore — Redis RESP2 store
  • Kotlin/Spring Boot distributed ledger
Get in touch

Let's build something that lasts.

Open to full-time roles and contract work, high-trust infrastructure and applied research especially.

jessica@templetsolutions.com LinkedIn GitHub