Show HN: NiceShot AI – The analytics layer competitive games forgot to add https://ift.tt/iPoqTO5

Show HN: NiceShot AI – The analytics layer competitive games forgot to add I've been working on NiceShot AI for the last 26 months as a way to experiment with computer vision for gameplay analysis. The basic idea is to take a recorded long gameplay session and automatically turn it into structured gameplay information and highlights. Bridging the performance analysis gap between match stats (too short) & lifetime stats (too long). The current pipeline is: Gameplay video → YOLO detection → event tracking → OCR/context filtering → event timestamps → clips → highlight compilation → session statistics For example, I can fine-tune a YOLO model to recognize game-specific HUD elements and then configure the pipeline to interpret those detections as events such as kills, deaths, or medals. One thing I wanted to avoid was making the entire system game-specific. Supporting another game should mostly require a new detector/model and configuration rather than rewriting the whole pipeline. And currently, I am working on that. I also use OCR for states that shouldn't be counted as gameplay events. For example, in Call of Duty games a kill indicator can appear during a "KillCam" or "Spectating" state, so the OCR layer can be used to filter those cases. The system can then produce clips around detected events and compile them into highlight reels, including vertical versions for Shorts/TikTok-style content. Recently, I have also been experimenting with using a lightweight Video LLM after event detection. Instead of sending the entire multi-hour gameplay session to the Video LLM model, I take the extracted short clips around an interesting event and ask the model to explain what happened. The goal is to eventually turn this into a lightweight conversational gameplay coach. The project currently runs locally on NVIDIA GPUs. Have experimented with an old GTX1650 4GB VRAM laptop and it works fine but slow ofcourse. Finally, I know that real-time event detection should be the way to go and this was my original goal, but I chose offline processing because I didn't want computer vision inference to interfere with the player's actual game performance. I'd also be interested in hearing how others would approach making the detector/inference layer fast enough for real-time gameplay without causing noticeable GPU/CPU overhead. Keeping in mind, I input 1080p frames into the model. The project is still evolving, so technical criticism and suggestions are very welcome. Thanks. https://ift.tt/nXKz5yA August 11, 2026 at 02:18AM

Show HN: I vibecoded a heavy lift drone https://ift.tt/cSdaze1

Show HN: I vibecoded a heavy lift drone https://ift.tt/2JBXnOE August 11, 2026 at 01:27AM

Show HN: Seedle, daily -dle for Minecraft seeds https://ift.tt/XWBQKFg

Show HN: Seedle, daily -dle for Minecraft seeds If you're not familiar with RNGdle, it's a daily game that randomly generates a 6 digit code. Pure dopamine delivery with little else of substance, but what else do you really need? I love the RNGdle idea, but I thought it might be even more fun as a way to discover Minecraft seeds. So I spent the day making this. For the uninitiated, Minecraft is deterministic, so you can determine what exactly the spawn point of a new world looks like if you know the seed number. This project gamifies that discovery process by assigning value to certain generated features, culminating in an overall score. It's just a little dopamine hit + small potential value for people who actually play Minecraft. https://ift.tt/gKvJYQq August 11, 2026 at 12:47AM

Show HN: Blueferry, iMessage on Linux https://ift.tt/PpoeYxD

Show HN: Blueferry, iMessage on Linux Hey; Blueferry is an app that connects to your iPhone via Bluetooth to expose an iMessage bridge to your Linux desktop. It supports receiving and sending text-based messages, either 1:1 or in group threads. It requires no phone-side software or jailbreak - just Bluetooth pairing. It accomplishes this by using an undocumented? behavior where enabling ANCS via Bluetooth LE (via https://ift.tt/4twUnik 's technique) causes the phone to accept message capability requests over standard Bluetooth. ancs4linux indicate that they've had mixed success with different BT chipsets on the Linux side, though all of the machines I've been able to test it on have been able to negotiate the connection. https://ift.tt/2rxBHIz August 10, 2026 at 01:35AM

Show HN: Voice driven murder mystery, Interview AI suspects with your voice https://ift.tt/QEFHd4T

Show HN: Voice driven murder mystery, Interview AI suspects with your voice Hey HN! I'm excited to show off this really fun project I put together. I originally built this project 2-3 years ago, AI was already booming at the time, however voice AI agents were still very early. I loved my proof of concept at the time, but wasn't quite happy with it. I recently had the desire to check out the tech again, and know many of you will be interested. Interviews are speech to speech with OpenAI's gpt-realtime-2.1 over WebRTC. This model is... expensive, and because of that, I have to add some amount of restrictions, conversations are tied to a authenticated Clerk user id. I have also added a 30 minute timer because well, I really don't want to go broke while I sleep tonight. Each suspect has a tool they call when you make a direct accusation. It captures who you accused and a faithful list of the evidence you actually stated. A separate gpt-5-mini judge then decides which of the case's required evidence facts you genuinely presented. Paraphrasing counts, vague suspicion and fishing don't. The rest is Next.js, MongoDB, and Clerk. Let me know whether the suspects hold up under a real interrogation. https://ift.tt/dgGbHWc August 10, 2026 at 12:18AM

Show HN: Gotcha- First on-device AI copilot for Android https://ift.tt/xo6GyBD

Show HN: Gotcha- First on-device AI copilot for Android https://ift.tt/ZrSOW6n August 9, 2026 at 11:34PM

Show HN: My WHOOP has no screen. My Kindle had no job. So I put them together https://ift.tt/bAH3JYa

Show HN: My WHOOP has no screen. My Kindle had no job. So I put them together I run a lot, so every morning I check my recovery and decide whether to run or take a rest day. I wanted an always-on screen that would let me see everything at a glance. Then I remembered the old Kindle collecting dust in the corner. So I built this: WHOOP API → a server on my Mac mini → grayscale PNG → Kindle screensaver. It’s now open source, including the full setup guide. https://ift.tt/jDcIR78 August 9, 2026 at 09:19AM