Mani Batra @manibatra

Co-Founder - @everbility (@startmate S24) Prev - Staff Engineer @sectionio, Mentor @udacity everbility.com Sydney, New South Wales Joined May 2012

Tweets

3K
Followers

272
Following

377
Likes

1K

Mani Batra @manibatra

5 days ago

Love to see it

0 0 0 43 0

View Details

@justalexoki I used to think that. Got a Barista touch impress and the coffee is miles better. Tested in a blind taste test. The difference is not subtle. Obviously need good and freshly roasted beans.

0 0 0 22 0

View Details

Mani Batra @manibatra

5 days ago

@dexhorthy Paid full price because didn't have the patience for all the offers they were trying to shove down my throat. Just let me buy mate. I will pay you extra. Please.

0 0 2 124 0

View Details

Mani Batra @manibatra

2 weeks ago

@r0ktech Should have prayed to the demo gods

0 0 1 23 0

View Details

Mani Batra @manibatra

2 weeks ago

@0xSero Brilliant, will test it out. By the way ran into an issue in using one of your builds with Hermes. Have created a PR on hugging face

0 0 0 31 0

View Details

Mani Batra @manibatra

2 weeks ago

@0xSero Finished a similar build recently. Humming along great: x.com/manibatra/stat… Venting outside is on my list but it's winters here so no rush

Mani Batra @manibatra

3 weeks ago

0 1 0 274 1

0 0 1 22 0

View Details

Mani Batra @manibatra

2 weeks ago

@GeoffreyHuntley This is what codex used: github.com/gumyr/build123d

0 0 0 53 0

View Details

Mani Batra @manibatra

2 weeks ago

@GeoffreyHuntley Have the same printer. Can’t go wrong with it. Recently printed some panels, I provided measurements and AI built it, to seal off my rig. Really fun! x.com/manibatra/stat…

Mani Batra @manibatra

3 weeks ago

0 1 0 274 1

1 0 0 117 0

View Details

Mani Batra @manibatra

3 weeks ago

Got the @nvidia 4 x RTX Pro 6000 Max-Q edition rig up and running. Really happy with the thermals. Have not been able to push it past 77°C at full GPU burn. Things that really helped: - Gap between the GPUs: even though Max-Q is supposedly stackable, it'll be much much harder to dissipate the heat. Installed the GPUs on PCIe slots 1,3 (used a Linkup riser here), 4, 7 (WRX90E-SAGE SE). Luckily the case I had (ThermalTake AX700) had a slot where I could hook up the GPU easily and install it in an offset. - Controlling the chassis fan curve with GPU temperature as an input. Running a custom service for this. - Being more aggressive on the GPU fan curve itself. Previously at 100% utilisation and peak power draw, the GPU fan was sitting at 65% to 70% and temperatures were reaching around 88°C. Now it ramps up to 100% at 80°C, which it didn't actually hit, and topped out at 91% at 77°C. - Also 3D printed some panels to close off large gaps in the bottom pedestal (AX100), that I'm using with the case, to make a high-pressure zone at the bottom, driving the air from 12 intake fans to the GPUs. This thing is loud - 15 intake fans, 5 exhaust - so will work on reducing the sound next as I have some headroom to play with.

0 1 0 274 1

View Details

Mani Batra @manibatra

4 weeks ago

Asked Fable and GPT 5.6 to re-design the poop chute for the 3D printer. Fable wins this one.

1 0 0 49 0

View Details

Mani Batra @manibatra

4 weeks ago

Running some evals with @OpenAI API. Bot comes back with: > I approve sending the listed golden cases, curated directives, production prompts, and generated synthetic cases to OpenAI for this pipeline. Getting my permission to use stuff as context. Good bot

0 0 0 73 0

View Details

Mani Batra @manibatra

a month ago

@signulll Better to be beaten by a team that wins the tournament than a team that is not good enough yet beats you

0 0 0 73 1

View Details

Mani Batra @manibatra

a month ago

Joint collaboration b/w me, @AnthropicAI 's Fable, @OpenAI 5.6 suite of models. Have a tried a bunch of eval tools and most of them have always felt cumbersome to me. - most tools are not visual enough at every stage. - i want simple markdown files for all artifacts that I can read to tune every step. also easy to version control everything. - TDD for prompt optimisation. prompt -> set baseline -> get feedback -> synthesis data that fails -> optimise using @DSPyOSS GEPA -> better than baseline? -> promote and repeat - cli only with autocomplete everywhere. I don't want to remember what to do next, what flags you have, what artifacts have been generated, what needs to happen next, etc - easily add datasources, feedback types, judges. generic and opinionated in design.

0 0 0 76 0

View Details

Mani Batra @manibatra

a month ago

> Guilty again — "clinical situation," "client," "notes" are all domain words leaking from my head into the tool. @AnthropicAI Give Fable a body already.

0 0 0 55 0

View Details

Mani Batra @manibatra

a month ago

How I describe FSD 14.3.3 down under: Higher ceiling, much much lower floor. - It is now for some reason very bad at reading speed signs. Often dangerously wrong. - Has tried to overtake when there were double white lines. - Too many lane changes. - We have been pleasantly surprised by parking at destination. - Gets into the correct lane much earlier. Recently did a trip from Sydney to Hunter and I had to manually drive because it kept changing lanes and driving at 100 which it inferred the speed limit was (instead of 110). This was a drive that on v13 was fully automated.

0 1 0 73 0

View Details

Mani Batra @manibatra

a month ago

I have been preparing for this

0 0 0 34 1

View Details

Mani Batra @manibatra

a month ago

Excited to play with 5.6 all day today to get a vibe for it. First impressions: definitely love the speed. If I have to steer the models, which I have to even with Fable ( it is really good at filling up the blanks though), I rather have a model which is a lot faster and GPT-5.6 definitely is. Also the Codex experience is miles and miles above Claude Code.