[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

# ![@deyneka_e Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::1374837001388707845.png) @deyneka_e Kate Deyneka

Kate Deyneka posts on X about infra, 6969, llms, tools for the most. They currently have XXXXX followers and XX posts still getting attention that total XXXXX engagements in the last XX hours.

### Engagements: XXXXX [#](/creator/twitter::1374837001388707845/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1374837001388707845/c:line/m:interactions.svg)

- X Week XXXXXX +15%
- X Month XXXXXXX +1,453%
- X Months XXXXXXX +6,367%
- X Year XXXXXXX +21,586%

### Mentions: XX [#](/creator/twitter::1374837001388707845/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1374837001388707845/c:line/m:posts_active.svg)

- X Week XX +83%
- X Month XX +209%
- X Months XX +2,367%
- X Year XX +7,500%

### Followers: XXXXX [#](/creator/twitter::1374837001388707845/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1374837001388707845/c:line/m:followers.svg)

- X Week XXXXX +7.20%
- X Month XXXXX +71%
- X Months XXXXX +296%
- X Year XXXXX +1,007%

### CreatorRank: XXXXXXX [#](/creator/twitter::1374837001388707845/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1374837001388707845/c:line/m:influencer_rank.svg)

### Social Influence [#](/creator/twitter::1374837001388707845/influence)
---

**Social category influence**
[technology brands](/list/technology-brands)  XXXXX% [vc firms](/list/vc-firms)  #81 [social networks](/list/social-networks)  XXXX% [stocks](/list/stocks)  XXXX%

**Social topic influence**
[infra](/topic/infra) #28, [6969](/topic/6969) 4.55%, [llms](/topic/llms) 4.55%, [tools for](/topic/tools-for) 4.55%, [snapchat](/topic/snapchat) 4.55%, [applied](/topic/applied) 4.55%, [futures](/topic/futures) 4.55%, [built in](/topic/built-in) 4.55%, [post it](/topic/post-it) 4.55%, [linkedin](/topic/linkedin) XXXX%

**Top accounts mentioned or mentioned by**
[@karpathy](/creator/undefined) [@dstackai](/creator/undefined) [@a16z](/creator/undefined) [@fal](/creator/undefined) [@dynamicwebpaige](/creator/undefined) [@potentialfung](/creator/undefined) [@liangsays](/creator/undefined) [@vargastartup](/creator/undefined) [@mlejva](/creator/undefined) [@weightsbiases](/creator/undefined) [@nestymee](/creator/undefined) [@speedrun](/creator/undefined) [@tszzl](/creator/undefined) [@bigthink](/creator/undefined) [@lumalabsai](/creator/undefined) [@elevenlabsio](/creator/undefined) [@theworldlabs](/creator/undefined) [@makodevai](/creator/undefined) [@scaleai](/creator/undefined) [@snowflake](/creator/undefined)
### Top Social Posts [#](/creator/twitter::1374837001388707845/posts)
---
Top posts by engagements in the last XX hours

"if you're looking for the perfect way to break into LLMs - this is it @karpathy released nanochat which is a great example of the entire LLM lifecycle: X. training the tokenizer X. pre-training on a large amount of diverse data X. finetuning on conversational data to adapt for chat use cases X. additional supervised finetuning (SFT) on higher-quality data + safety X. reinforcement learning to squeeze out a few more percentage points of performance additionally youll learn how to log everything with @weights_biases - because logging and monitoring are essential for model training"  
[X Link](https://x.com/deyneka_e/status/1977888930365407273) [@deyneka_e](/creator/x/deyneka_e) 2025-10-14T00:07Z 3396 followers, 125.1K engagements


"sunday night - getting ready to speak at the AI infra & Open-Source meetup by @dstackai. Ill be talking about methods to optimize video generation inference. a thread on that is coming soon too"  
[X Link](https://x.com/deyneka_e/status/1980143851496161361) [@deyneka_e](/creator/x/deyneka_e) 2025-10-20T05:27Z 3395 followers, 4927 engagements


"@nestymee @speedrun ahaha a16z is responsible for heart attacks"  
[X Link](https://x.com/deyneka_e/status/1981081651200610326) [@deyneka_e](/creator/x/deyneka_e) 2025-10-22T19:34Z 3396 followers, XXX engagements


"I left my full-time job to build stuff. I used to be a Senior ML Engineer at Snapchat focusing on Video Generation - both applied research and product. Loved my job but its time to move on. Right now feels like the best time to build. Ill start with small and medium AI apps and tools (some just for fun). My bigger interest is in dev tools for research. If you want to chat about any of this my DMs are open Some things I believe about building as of fall 2025 (thread)"  
[X Link](https://x.com/deyneka_e/status/1977770548135702759) [@deyneka_e](/creator/x/deyneka_e) 2025-10-13T16:16Z 3396 followers, 55.1K engagements


"just move to sf and it will change your life"  
[X Link](https://x.com/deyneka_e/status/1980445666343284916) [@deyneka_e](/creator/x/deyneka_e) 2025-10-21T01:26Z 3396 followers, 64.5K engagements


"this is an amazing work congrats @a16z new media team👏🏻"  
[X Link](https://x.com/deyneka_e/status/1981122751156093360) [@deyneka_e](/creator/x/deyneka_e) 2025-10-22T22:17Z 3396 followers, 3048 engagements


"2. No need to raise on an idea. With todays AI tools an MVP can be built in a week. Post it on X IG LinkedIn Reddit - get traction maybe even paying users then raise. Its better to bootstrap at first especially if you have savings to cover your basics"  
[X Link](https://x.com/deyneka_e/status/1977770554120908803) [@deyneka_e](/creator/x/deyneka_e) 2025-10-13T16:16Z 3387 followers, 2185 engagements


"3. Skip training and fine-tuning models early on. If youre building a product (not a model) theres no need to train or fine-tune at the MVP stage. Current models are already powerful and good prompt engineering can take you far. Using APIs also saves costs on model deployment and infrastructure"  
[X Link](https://x.com/deyneka_e/status/1977770555857306090) [@deyneka_e](/creator/x/deyneka_e) 2025-10-13T16:17Z 3387 followers, 2151 engagements


"me watching Ramp ads"  
[X Link](https://x.com/deyneka_e/status/1978322138651045926) [@deyneka_e](/creator/x/deyneka_e) 2025-10-15T04:48Z 3380 followers, 45.5K engagements


"now that everyones been kicked out of sf is it just me and the AI agents left"  
[X Link](https://x.com/deyneka_e/status/1978876901784727563) [@deyneka_e](/creator/x/deyneka_e) 2025-10-16T17:33Z 3392 followers, 5168 engagements


"Aura Night with @fal and @DynamicWebPaige ✨🍌🔮"  
[X Link](https://x.com/deyneka_e/status/1979035306247548980) [@deyneka_e](/creator/x/deyneka_e) 2025-10-17T04:02Z 3396 followers, 1520 engagements


"@potentialfung @fal @DynamicWebPaige about generative art and media"  
[X Link](https://x.com/deyneka_e/status/1979044008807575986) [@deyneka_e](/creator/x/deyneka_e) 2025-10-17T04:37Z 3396 followers, XX engagements


"Ill be speaking this Tuesday at the AI infra & open-source meetup. come join"  
[X Link](https://x.com/deyneka_e/status/1979283668838580350) [@deyneka_e](/creator/x/deyneka_e) 2025-10-17T20:29Z 3392 followers, 2396 engagements


"@liangsays @a16z looking forward sooo much"  
[X Link](https://x.com/deyneka_e/status/1980340851177071094) [@deyneka_e](/creator/x/deyneka_e) 2025-10-20T18:30Z 3398 followers, XXX engagements


"@vargastartup wow interesting. whats the model indeed I feel that the majority of the models add safety at the level of the prompt rewriting and the data can still be biased. its hard to get good video data in general and how to get good unbiased video data - even more challenging"  
[X Link](https://x.com/deyneka_e/status/1980359752787075367) [@deyneka_e](/creator/x/deyneka_e) 2025-10-20T19:45Z 3380 followers, XXX engagements


"@WilcoKr @karpathy that's interesting which editors do you prefer and why don'y you like Cursor/VSCode debugger"  
[X Link](https://x.com/deyneka_e/status/1980676562145145208) [@deyneka_e](/creator/x/deyneka_e) 2025-10-21T16:44Z 3395 followers, XX engagements


"Video generation is slow. X techniques to make your model 10-40x faster I spoke yesterday at the AI Infra & Open-Source Meetup by @dstackai about how to make video models more efficient. Here are the main highlights - with all the links you need for faster video diffusion inference"  
[X Link](https://x.com/deyneka_e/status/1981086482195763430) [@deyneka_e](/creator/x/deyneka_e) 2025-10-22T19:53Z 3396 followers, 3691 engagements


"hot take: nyc should compete with nvidia"  
[X Link](https://x.com/deyneka_e/status/1981433474352632270) [@deyneka_e](/creator/x/deyneka_e) 2025-10-23T18:52Z 3396 followers, 1625 engagements