Deepak Mishra (@deepak_graphon) header
Deepak Mishra (@deepak_graphon) profile picture

Co-founder @GraphonAI, the context and memory layer for enterprise AI. Sharing what enterprise customers actually say about putting AI to work

San Francisco, CA
Joined January 2022
Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)Photo by Deepak Mishra (@deepak_graphon)
Deepak Mishra retweeted
I mapped an entire movie by mood for about 5 cents (Jev @typesafeai + @GraphonAI). Moods like: Funny, Scary, Thrilling, Suspenseful etc. Around 7,000 AI analyses for roughly $0.05 and 7 seconds. The demo is pretty simple: First, Graphon maps the movie. We ask Graphon about 40 questions designed to cover what's happening throughout the video. Graphon returns descriptions of the scenes, along with citations to the exact moments in the movie. We use those questions to build a picture of what's happening across the entire movie. Then we send those scene descriptions to Jev from @TypeSafeai Jev looks at each scene and judges the probability that it is funny, scary, suspenseful, exciting, calm, or whatever other categories we want to analyze. That's where we do around 7,000 analyses in parallel. The result is basically a mood map of the entire movie. Now imagine you're a movie studio, video editor, or creator. You could say: “Show me every scary moment.” “Find all the car chases.” “Give me the funniest 30 seconds.” “Create a compilation of the most suspenseful moments.” And jump directly to those exact scenes. Today, doing this at scale is surprisingly complicated. Someone has to watch, tag, classify, scrub through footage, and manually find all of these moments. With Graphon + Jev, you can start doing it automatically. And the crazy part? Those ~7,000 mood analyses cost about 5 cents. This is what I think multimodal search starts to look like when you combine understanding, citations, and reasoning across an entire video. @CompleteSkeptic what do you think?
3
2
1
3
584
Deepak Mishra retweeted
At Graphon, we believe intelligence lives in the connections, not the nodes. So we are bringing the graph to the table. We are hosting our first Engineering Leaders Dinner sponsored by J.P. Morgan with confirmed attendees from OpenAI, Anthropic, Harvey, Carta, Google, Hippocratic, Lucid Motors, Prime Intellect, Mercor and more. We have a few seats left and if you are passionate about topics like context intelligence, memory, knowledge graphs, multimodal data and LLMs would love to have you join 👇 luma.com/k7nnf5g6 #MultimodalAI #EngineeringLeaders #Agents #ContextIntelligence #Memory #Graphon
1
2
408
Deepak Mishra retweeted
We talked to dozens of Fortune 500 AI leaders while building Graphon. One thing has become clear: enterprise AI doesn’t need bigger models, it needs better architecture. Systems built for text break under multimodal complexity. Teams aren’t leaving for marginal gains; they’re leaving because current architectures can’t keep up. A quiet inflection point is here. CC: @ArbaazK87777726, @deepakmishraVC
1
2
187
Deepak Mishra retweeted
🚨New Paper Alert! 🚨 Our ICML 2025 paper (led by @Narmeen29013644) shows how small models can help steer and align much larger ones by building “bridges” between them.🧵👇
5
10
12
43
7,004
Hackers & Backers hosted another remarkable event, this time on Data Center Energy Management. Here are a few key takeaways that stood out:
1
1
3
400
Deepak Mishra retweeted
OpenAI's top safety researchers quit last week. It's easy to say this is OpenAI's fault, but it's not -- it's incentives. Capitalism and safety aren’t aligned. And capitalism always wins. To make progress on safety, we need to align those incentives. Here’s how we do it 🧵
2
9
1
26
4,104
Deepak Mishra retweeted
Neural networks are notoriously opaque. Let’s change that. At Martian, we’re developing methods to make black-box models transparent, editable, and verifiable. Called “model mapping”, it powers 1K+ businesses through our model router. A sneak peek of what we’re working on:
3
12
4
62
13,149
Deepak Mishra retweeted
🚀 Introducing RouterBench, the first comprehensive benchmark for evaluating LLM routers! 🎉 A collaboration between @withmartian and Prof. @KurtKeutzer at @UCBerkeley, we've created the first holistic framework to assess LLM routing systems. 🧵1/8 To read more: blog.withmartian.com/post/ro…
8
29
13
129
50,950
Deepak Mishra retweeted
Some companies paid $40K for a booth at #Nvidia #GTC24. We paid $500 to rent a Cybertruck and got way more leads. If you're not using our #ModelRouter in your AI application, you're overpaying too: withmartian.com
9
13
4
70
8,947
Happy republic day to India 🇮🇳! Every year I make a post on this day and usually talk about how well India is doing on so many economic indicators. This year, I will do something different. Over the last few years, it's become clear to the world that India is on the ascend and will play an increasingly important role in an increasingly multipolar world. We are showing up on the front-foot everywhere - G20, Davos, Asian games, Cricket, Bollywood, Technology, Infrastructure, Space, Defense, the list is very long. Internally, the changes are very visible too. Our airports used to look like matchboxes and now they win global design awards Our trains used to look boring and outmoded, and now the fleet is getting quickly modernized. Our infrastructure spending is the highest it's been in decades. India, the nation, is at its most confident. A sign of a confident nation is that we can look at all the amazing things we've achieved and say 'yeah, that's great, but we've got so much work yet to do' and then get back to it. So, this year I want to talk about things we need to fix in this decade if we are to continue growing for the next few decades and chart a path to the top 3 most powerful countries in the world. 👇
7
25
2
196
32,467
Deepak Mishra retweeted
🚀Introducing The LLM Inference Provider Leaderboard leaderboard.withmartian.com - a live-updated, unbiased eval of API Inference products. Featuring: @abacusai, @anyscalecompute, @DeepInfra, @DecartAI, @FireworksAI_HQ, @LeptonAI, @togethercompute, @perplexity_ai, @replicate, as well as @OpenAI and @AnthropicAI models For each provider's Mixtral-8x7B and Llama-2-70B-Chat public endpoint, we benchmark cost, rate limit, P50 & P90 of throughput & TTFT, and average daily collections overtime for long term tracking. At Martian, we route each API request to the best LLM to reduce cost, reduce latency, and get the best performance. So finding the best providers is an important problem for us. We found that there's a > 5x cost difference, > 6x throughput variation, and even larger rate limit discrepancies among providers! Choosing between different LLMs is only part of the equation -- the selection of different inference endpoints is also crucial to get the best performance for your use case. See highlights of provider performance in🧵👇
8
55
11
266
129,052
The last Hackers and Backers event of 2023 wrapped up with holiday cheer 🚀 We've grown to 200+ members, learned about numerous topics, shared deal flow, and achieved fundraising and hiring successes. Excited for 2024 holds - reach out to learn more #Community #Learning
1
2
261
Deepak Mishra retweeted
We are thrilled to share that Deepak Mishra at Avanta Ventures has been named in this year's Rising Star Pre-Partner Venture Investors list. Congratulations on this well-deserved recognition! #AlphaPartners #AvantaVentures #VC #RisingStars alphapartners.com/news/2023/…
1
1
2
474
Selling B2B AI Infrastructure platform to banks/insurers? Understand their pain points, quantify value, focus on use case, highlight security, and offer pilots. Speak their language first - outline outcomes not features. Persistence pays off with long sales cycles!
1
1
1
251
Our yearly Napa tradition continues! Our group featured founders & VCs from e-commerce, SMBs, digital health, climate, biotech, fintech, insurtech, gaming, customer value & more - representing diverse industries & thought-provoking perspectives.
4
14
982
Clear metrics are HUGELY underrated in Generative AI. There are lots of cool demos coming out that turn into at best buggy products. A founder recently shared …..
2
1
2
315
We are excited to announce that our community of over 120 members has made the decision to rebrand BrunchNLearn to Hackers and Backers 🙏🎉🚀
3
18
634