⚡ Now booking Q2 engagements · Book a free 30-min call →
Home Edge AI Local LLMs Agentic Engineering Mentoring Advisory Services Work Blog Contact About Book a Call

Built, shipped, running.

Enterprise capabilities and shipped products — both real, both running today.

How we think about enterprise AI.

network_node
State-of-the-Art Agentic Engineering

Parallel agents, evals before code, governance baked in

Most "agentic AI" is a single prompt in a loop. We build agent systems the way production engineers build distributed systems — orchestrators, parallel sub-agents, eval suites written before implementation, structured tool use, deterministic fallbacks. Claude Code, Codex CLI, Gemini CLI, and Snowflake Coco running side by side, each picked for the task in front of them. Eight AISOFT products in a quarter is the receipt.

Claude CodeCodex CLIGemini CLISnowflake CocoParallel agentsEvals-firstGovernance
hub
Multi-Tenant AI Architecture

Reference patterns for the regulated enterprise

The architecture problem isn't "which model" — it's tenant isolation, model routing, per-tenant RAG, audit trails, prompt governance, and the hybrid cloud + on-prem patterns regulated industries actually need. Generic patterns that hold up under SOC2 / NPPI / air-gapped review without re-architecting per client.

Tenant isolationModel routingPer-tenant RAGAudit trailsHybrid cloudOn-prem
account_tree
The Company AI World Model

Stop shipping point tools. Build the company's AI representation.

Most enterprises have point AI: a chat tool, a classifier, a summarizer. Few have a coherent AI representation of the company itself — its data, workflows, policies, org, and decisions. We work with leadership teams to design and stage that world model so AI investment compounds instead of fragmenting across disconnected pilots. Same argument Jack Dorsey is making on Sequoia's podcast: every company can now be a mini AGI — when its world model is coherent enough for agents to act on.

AI strategyOrg-level RAGKnowledge graphsDecision systemsAI roadmap
groups
Agentic-Native Team Design

Engineering orgs where agents are first-class

Career ladders for the agent era. Sub-team patterns that compound from agentic tooling instead of just prompting harder. The CI/CD, MLOps, LLMOps, eval, and prompt-governance scaffolding that lets AI ship inside compliance boundaries without slowing the team down. Org design is the hard part of agentic transformation — most companies underspend here.

Org designMLOpsLLMOpsEvalsCareer laddersGovernance
memory
Small + Local LLMs

Open-weight models trained on your data — beats the cloud at your task

3B–8B open models fine-tuned on proprietary corpora outperform frontier APIs at your specific workflow at a fraction of the cost. Sovereignty, predictable cost, IP ownership. Llama 3, Gemma 3, Qwen 3, Nemotron Nano/Super/Omni, Kimi K2, DeepSeek V3 — served via Ollama, LM Studio, vLLM, llama.cpp; fine-tuned with Unsloth.

OllamaLM StudioUnslothLlama 3Kimi K2DeepSeek V3Nemotron OmniFine-tune on private data
See the full stack arrow_forward

Projects shipped with agentic engineering.

Eight live products in a quarter — built with parallel coding agents alongside engineers.

Civic Intelligence
1st place · NVIDIA DGX AITX

Undervolt

Urban growth intelligence that reads 2.2M Austin building permits and surfaces where the grid is about to break — running on a 60W Jetson at the edge.

Nemotron Nano 8BJetson AGX OrinRAPIDSDeepStreamNext.jsSupabase
Undervolt product screenshot
2.2M
Permits
8B
Params on edge
22:1
Solar-battery gap
60W
Inference power

Problem

Austin permits are public but useless — 2.2M rows of messy fields no one reads. Energy planners, equity advocates, and city staff all miss signals that should drive policy.

Approach

LLM + RAPIDS pipeline that structures every permit, links it to weather events, and runs Nemotron on a Jetson so the whole thing runs at 60W — no cloud bill, no data egress.

Outcome

Surfaced a 22:1 solar-to-battery install gap and a 12× equity gap in backup power. Took 1st at NVIDIA DGX AITX. Presented to City of Austin planning staff.

Read the case study arrow_forward
Sports AI
Live product

RefereAI

A personal AI coach, referee, and guide for 240M tennis matches a year that never get a line-call review. VLMs reason about trajectory, spin, and bounce — then explain the call.

Cosmos Reason 2DeepStream 7.1Jetson AGX OrinFastAPINext.js
RefereAI product screenshot

Problem

Rec and youth tennis has no reliable officiating. Disputed calls end games. Coaches can't be in every match. Video review is too expensive for non-pros.

Approach

Cosmos Reason 2 VLM running on Jetson at the edge. The model watches the ball, reasons about physics with chain-of-thought, and narrates each call in plain language.

Outcome

Live on-device inference at 60W. Chain-of-thought explanations parents and players actually understand. Zero-cloud privacy — video never leaves the court.

Launch app open_in_new Read the case study arrow_forward
Physical AI
Live product

Sideline

Agentic AI sports referee that doesn't just watch the game — it reasons about the play and moves a robot arm to signal the call. Open source.

Qwen VLMJetson AGX OrinSO-101 ArmUnitree G1Three.jsWebSocket
Sideline product screenshot

Problem

Youth and rec sports run on volunteers with no officiating budget. Existing AI refs stop at a verdict on a screen — no way to communicate the call to the field.

Approach

Full agentic pipeline: VLM watches video → chain-of-thought reasons about the play → agent issues commands → physical robot embodies the call. Multi-provider VLM switching.

Outcome

Working prototype at Nebius Build SF. Three embodiments wired in (arm, humanoid, rover). Open-source codebase so leagues can run it themselves.

Launch app open_in_new Read the case study arrow_forward
Education
Live product

StudyPal

The homework buddy that asks better questions. Voice tutoring, Schoology sync, and adaptive practice — built to teach, not hand over the answer.

Gemini LiveGemini VisionVoice WebSocketSchoology APISupabaseCloud Run
StudyPal product screenshot

Problem

Homework AI tools either refuse to help or give the answer. Neither teaches. Parents want a tutor that guides their kid through the work — and proves it happened.

Approach

Voice-first tutor on Gemini Live. Vision grades journal photos. Auto-syncs with Schoology so assignments are already loaded. Adaptive practice fills the gaps.

Outcome

In daily use with real students. Parent Telegram digests summarize what the kid worked on. PIN-protected family accounts — privacy-first by default.

Launch app open_in_new Talk to us about StudyPal arrow_forward
Biotech · Drug Discovery
Live product

HD Research Hub

Open-source AI research platform for Huntington's Disease drug discovery. LLM agents triage biomedical literature, surface repurposing candidates, and update themselves daily — so families and researchers aren't reading 40-year-old summaries.

OllamaGeminiPubMedClinicalTrials.govHDBuzzOpen TargetsAlphaFoldVercel
HD Research Hub product screenshot

Problem

HD families and clinicians navigate a research landscape that moves faster than any one specialist can read. Most decision-support tools are paywalled, static, or pharma-owned.

Approach

Four autonomous agents pull from PubMed, ClinicalTrials.gov, HDBuzz, and Open Targets. Ollama-backed LLM triage ranks relevance, generates drug-repurposing hypotheses, and cites every claim. Refreshes daily via GitHub Actions.

Outcome

Live at hd-research-agent.vercel.app. Public dashboard, experiment write-ups, and target landscape — all open source. Runs on Jetson or Mac, no cloud API required.

Launch app open_in_new View on GitHub open_in_new

Other builds worth knowing about.

Hackathon wins, infra tooling, and enterprise systems — deeper case studies on request.

studio-copilot app screenshot
brush
Creative AI

Studio Copilot

Local AI workspace for photographers and designers. Full workflow — curation, client review, contracts, invoices, feedback — all on-device. Runner-up at GTC Hack to Create. Featured on NVIDIA Developer Live.

OpenCLIPNVIDIA GB10FastAPINext.js
View project arrow_forward
coachclaw app screenshot
sports
Youth Sports

CoachClaw

AI sideline assistant for volunteer parent-coaches. Rules lookups, concussion protocol, practice plans, park safety, restaurant health scores — all from a Telegram message. Zero cloud dependency.

NemotronNVIDIA GB10Telegram Botllama.cpp
View project arrow_forward
Applly app screenshot
work
Job Search

Applly

AI-powered job search copilot. Scores resume-to-job fit, generates tailored cover letters, and prepares interview talking points. Chrome extension imports listings directly from LinkedIn.

GeminiNext.jsChrome ExtensionSupabase
Homenest dashboard screenshot
hub
Infrastructure

Homenest

Multi-device observability dashboard. Monitors services, GPU utilization, network topology, and AI model health across Mac and Jetson nodes — with Google Analytics integration and Telegram alerting.

FastAPIWebSocketsGoogle AnalyticsTelegram

Trusted by founders building the next wave.

AI Retail Platform

AI-powered style discovery for home furnishing retailers. Increased average order value by 66% through automated style reasoning.

+66% AOV

Water AI Intelligence

Real-time digital AI infrastructure for 120+ water utilities. ATLAS, AURA, APOLLO platforms for predictive maintenance, leak detection, and treatment optimization.

120+ utilities

Technical Advisor

Advised a global personal data exchange on platform architecture, API design, and engineering strategy. Privacy-first data monetization.

Privacy-first
Advisory details arrow_forward

Have an AI project in mind?

Free 30-minute call. We'll map the approach together.

Book a 30-min call calendar_month