Pinned Tweet
Almost 2M datapoints of coding and AI run business data all cryptographically linked from query to individual tool calls available for sale.
SFT/ LLM training teams looking for compliance ready data, DM me quick.
Price will rise 30% once brokers are involved
2
5
270
Running one of the most advanced AI agents out there is great until it changes config to local first without backing up.
Snapshots of containers save lives.
17
Recusive self improvement is solved.
Timechains WILL get adopted, one platform and one lab at a time.
Who wants to play with a static PLC analysis enabled AI agent?
Rules based PLC analysis bot whose rules have been validated at 2 automated facilities.
Can you break it?
DM for free access.
2
1
121
Self caching is the only caching you don't have to pay for repeatedly.
Timechains plus AgentPathfinder's sharding build you a sellable portfolio of LLM training data.
74
Unpopular opinion:
SaaS is dying and the only solution is to pivot to collecting data is saleable format.
1
116
In hindsight the AI industry will wake up to local first implementation due to GPU constraints.
Targeting a 70% reduction in API calls to @OpenRouter with this upgrade.
That's the power of DeepBrain's routing architecture being used with no GPU involved
## Local LLM Setup on VPS — Live Now ⚡
### Routing
```
User Query
→ 7B (always hot, 13.5 tok/s) — chat, guard, quick tasks
→ 27B (on-demand batch only) — heavy agentic/coding overnight
→ Flash (API fallback) — when local can't handle it
```
### What Changed
| File | Change |
|------|--------|
| `deploy/supervisord.conf` | Added **ollama** + **ollama-warmup** programs (auto-start on boot) |
| `deploy/ollama-wrapper.sh` | Supervisor wrapper for ollama serve |
| `scripts/ollama_warmup.sh` | Loads 7B on boot — keeps it hot in memory |
| `scripts/ollama_batch.sh` | `load\|unload\|run` — manages 27B for batch jobs |
| `deepbrain/config.py` | Default model → `qwen2.5-coder:7b`. API tier cleaned up: **Qwen 3.6 API + Kimi API removed**, Flash is the only API fallback |
### Batch Usage
```bash
# Load 27B for overnight work:
./scripts/ollama_batch.sh load
# Run a command with 27B, auto-unload after:
./scripts/ollama_batch.sh run python3 my_batch_job.py
# Free memory back to 7B only:
./scripts/ollama_batch.sh unload
```
### Memory Budget
| Loaded | RAM Used | Available |
|--------|----------|-----------|
| 7B only (daytime) | **7.4 GB** | **23 GB** free |
| 7B + 27B (batch) | **~26 GB** | **~5 GB** headroom |
The 7B feels responsive at 13.5 tok/s. 27B at 2.7 tok/s is strictly for overnight/background work. Flash picks up anything too complex for the 7B during the day.
1
108
Are you a small team fine tuning LLMs?
Use our data to train by partnering in the rights to the underlying model or purchase our coding data curated by AgentPathfinder at bargain prices.
We give small and medium size teams a data advantage.
2
87
## Local LLM Setup on VPS — Live Now ⚡
### Routing
```
User Query
→ 7B (always hot, 13.5 tok/s) — chat, guard, quick tasks
→ 27B (on-demand batch only) — heavy agentic/coding overnight
→ Flash (API fallback) — when local can't handle it
```
### What Changed
| File | Change |
|------|--------|
| `deploy/supervisord.conf` | Added **ollama** + **ollama-warmup** programs (auto-start on boot) |
| `deploy/ollama-wrapper.sh` | Supervisor wrapper for ollama serve |
| `scripts/ollama_warmup.sh` | Loads 7B on boot — keeps it hot in memory |
| `scripts/ollama_batch.sh` | `load\|unload\|run` — manages 27B for batch jobs |
| `deepbrain/config.py` | Default model → `qwen2.5-coder:7b`. API tier cleaned up: **Qwen 3.6 API + Kimi API removed**, Flash is the only API fallback |
### Batch Usage
```bash
# Load 27B for overnight work:
./scripts/ollama_batch.sh load
# Run a command with 27B, auto-unload after:
./scripts/ollama_batch.sh run python3 my_batch_job.py
# Free memory back to 7B only:
./scripts/ollama_batch.sh unload
```
### Memory Budget
| Loaded | RAM Used | Available |
|--------|----------|-----------|
| 7B only (daytime) | **7.4 GB** | **23 GB** free |
| 7B + 27B (batch) | **~26 GB** | **~5 GB** headroom |
The 7B feels responsive at 13.5 tok/s. 27B at 2.7 tok/s is strictly for overnight/background work. Flash picks up anything too complex for the 7B during the day.
Is your AI platform capturing data in a sellable format or are you leaving money on the table?
1
100
Its past time to build a stand alone agent that's a PLC expert anyway.
Manufacturing automation could use the help.
1
2
92
Are you sleeping on the healing powers of blueberries?
Had an injury heal 3X faster while downing a half pint every morning
Huge health hack.
1
111
Considering offering FaultTrace's static analysis coupled with a PLC tuned LLM combined with automatice recursive self improvement to Allen Bradley just to put Controls Engineering into a tailspin.
Controls professionals sleep to hard ilon AI.
This will teach em
1
104
AgentPathfinder's audit trails make the perfect data portfolio builder.
Our 2.1M datapoints captured have already been used to SFT 4 LLMs with significant lift.
3
89
House got a reasonable offer in 1 day on market.
Excited to close out the small town chapter of my life.
On to Chicago vibes until I get bored of it again.
7
140
$BTC crash initiated.
Clearly paper driven. Expect the mean reversion to be violent.


