funnel: add renezander.com + Upwork callouts (top + Work-with-me section)
Browse files
README.md
CHANGED
|
@@ -34,6 +34,15 @@ datasets:
|
|
| 34 |
|
| 35 |
> **The local UI-grounding specialist for hybrid AI agents.** Drop in a screenshot + text target, get a strict JSON bbox. 2B params. MLX-native. Apache 2.0.
|
| 36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
## Why this exists — the hybrid AI argument
|
| 38 |
|
| 39 |
Today, most AI agents route **every** screenshot to a cloud frontier model (GPT-4V, Claude Vision, Gemini) just to find click coordinates. That's a $0.01–0.05 multimodal call adding 800ms–2s of latency, repeated 20–50× per agent run. Cost and latency compound. Screenshots full of private UI leave your machine.
|
|
@@ -163,6 +172,22 @@ Full training scripts (private repo, request access): [renezander030/imgparse-ti
|
|
| 163 |
- **Custom agent stacks** that need a $0/call grounding step instead of GPT-4V per screenshot
|
| 164 |
- **Self-hosted compound-AI systems** with a routing layer (specialist model for grounding, general LLM for planning)
|
| 165 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 166 |
## Citation
|
| 167 |
|
| 168 |
```bibtex
|
|
|
|
| 34 |
|
| 35 |
> **The local UI-grounding specialist for hybrid AI agents.** Drop in a screenshot + text target, get a strict JSON bbox. 2B params. MLX-native. Apache 2.0.
|
| 36 |
|
| 37 |
+
---
|
| 38 |
+
|
| 39 |
+
> **Want a specialist local model for *your* agent stack?**
|
| 40 |
+
> Built by **Rene Zander**, freelance AI engineer (DE/EN, remote). Custom fine-tunes, hybrid-AI architectures, on-prem deployments.
|
| 41 |
+
> → Hire directly on **[Upwork](https://www.upwork.com/freelancers/reneza)**
|
| 42 |
+
> → Or reach out via **[renezander.com](https://renezander.com)**
|
| 43 |
+
|
| 44 |
+
---
|
| 45 |
+
|
| 46 |
## Why this exists — the hybrid AI argument
|
| 47 |
|
| 48 |
Today, most AI agents route **every** screenshot to a cloud frontier model (GPT-4V, Claude Vision, Gemini) just to find click coordinates. That's a $0.01–0.05 multimodal call adding 800ms–2s of latency, repeated 20–50× per agent run. Cost and latency compound. Screenshots full of private UI leave your machine.
|
|
|
|
| 172 |
- **Custom agent stacks** that need a $0/call grounding step instead of GPT-4V per screenshot
|
| 173 |
- **Self-hosted compound-AI systems** with a routing layer (specialist model for grounding, general LLM for planning)
|
| 174 |
|
| 175 |
+
## Work with me
|
| 176 |
+
|
| 177 |
+
This adapter is a public reference of the recipe I deliver to freelance clients: small, fast, structured-output local specialists that slot into compound-AI agent stacks and cut cloud-LLM bills without losing capability.
|
| 178 |
+
|
| 179 |
+
If you need one of these, I can build it:
|
| 180 |
+
|
| 181 |
+
- a **UI-grounding model trained on your own product's screenshots** — your dashboard, your app, your customer interfaces — for higher recall on the elements your agents actually click
|
| 182 |
+
- a **hybrid agent architecture** that routes narrow tasks (grounding, OCR, classification, embedding, extraction) to local specialist models and reserves cloud frontier LLMs for the reasoning that actually needs them
|
| 183 |
+
- an **on-prem agent deployment** — Apple Silicon (MLX), CUDA box, or your existing K8s — with no screenshots leaving your infrastructure
|
| 184 |
+
- a **structured-output evaluation harness** that tells you when the local model is actually good enough to replace the cloud call in production
|
| 185 |
+
|
| 186 |
+
**Two ways to engage:**
|
| 187 |
+
|
| 188 |
+
- **Upwork** — contract-ready, vetted, pay-as-you-go: <https://www.upwork.com/freelancers/reneza>
|
| 189 |
+
- **Direct** — for longer engagements, retainers, or a quick conversation: <https://renezander.com>
|
| 190 |
+
|
| 191 |
## Citation
|
| 192 |
|
| 193 |
```bibtex
|