work

Projects

Five projects, five hard problems. Each one built to prove something real about systems design, AI evaluation, agent reliability, or distributed computing.

01
publicagent systems

Archon

Autonomous agent in Python — planner/executor/reflector with a typed middleware chain (schema validation, retries, tracing on every tool call) and a statistical evaluation harness: seeded runs, per-step traces, and a failure-mode taxonomy.

PythonAgentsEval
read case study
02
public · MITapplied ai

Sentinel

AI code review with hybrid retrieval and a deterministic eval harness.

FastAPIpgvectorBM25PythonNext.jsPostgreSQL
read case study
03
working study · MITmulti-tenant saas2026

Kairos

Multi-tenant OKR tracker that enforces tenant isolation in PostgreSQL with Row-Level Security, not in application code.

Spring Boot 3.3PostgreSQL 16Redis 7Java 21Next.js 14OpenTelemetry
read case study
04
public · single-node stabledistributed systems

Helios

Time-series database in Go — built from first principles.

GoLSM-treeGorillaWALPromQLRaft (optional)
read case study
05
public · MITml research

NeuroLens

MVP adversarial-ML research: a cross-modal CLIP → ResNet transfer study.

PyTorchCLIP-litePythonFGSMPGD
read case study