Where the AI market is talking

Ask HN: Who wants to be hired? (July 2026)

Location: Rawalpindi, Pakistan Remote: Yes Willing to relocate: Yes if assisted. Technologies: LangGraph, RAG, FastAPI, Qdrant, AWS Bedrock, LLM agentic pipelines Résumé/CV: muhammadsamar.vercel.app Email: samarshehzad598@gmail.com

Show HN: I built an AI agent to yell at me about my ADHD

Thanks! You made me realize I completely forgot to mention memory in my post . For running memory, hex maintains a 10 message full-length context window, with the rest being summarized when the threshold is being crossed (15 messages) by a separate LLM call and stored in vector storage (Qdrant). Aside from that, observational memories get constantly stored in Qdrant as well for convenient retrieval and embeddings are calculated locally via Qdrant's FastEmbed. Preferences, facts and things like that are store in SQLite.

Ask HN: Who wants to be hired? (July 2026)

Location: Oslo, Norway (CET) Remote: Yes (EU/US time zones) Willing to relocate: Yes. Technologies: Rust, Python (PyTorch, NumPy/Pandas, Flask, Django), PyO3 Databases/infra: Postgres, ClickHouse, Qdrant, Redis, Kafka, Docker, Kubernetes (k8s) ML and systems: vLLM, TensorRT, Ray, edge ML inference, gRPC/Protobuf, systems profiling (perf, flamegraphs, Valgrind) Rust focus: Async services (Tokio, Axum/Actix), data/compute (Polars/Arrow), database proxies, performance-critical pipelines. Domain expertise: - Academic: PhD in mathematics (topological data analysis, representation theory), MSc finance, MSc neuroscience. - Applied: database proxies/networking, quant dev and backtesting engines, ML/DL for banking & finance (risk/fraud), LLMs and a…

Ask HN: Who is hiring? (July 2026)

Starbridge | Senior AI Engineer (Python) | NYC or Remote | Full-time | Starbridge is building an AI platform that turns large-scale public and enterprise data into reliable sales insights. We are early, moving fast, and building from zero to one, so this role will have huge ownership and product impact. AI Engineer: Applied AI plus strong software engineering. You will build, evaluate, and deploy LLM-driven features like deep document analysis and interactive chat, working with models from OpenAI, Anthropic, and Gemini. Expect hands-on Python, ML system design, experimentation, and production reliability. Bonus for RAG depth and frameworks like LangChain, LlamaIndex, or Hugging Face. We're looking to build our in-person team in NYC but also open to remote! Apply: https://starbri…

Show HN: PulseDB – Unified Redis and Pinecone Alternative in Python

Show HN: Shikhu – Understand the code your agents write

Agents write code really well, but it can get hard to understand what they've written and why. Shikhu is a CLI tool and Agent Skill that facilitates learning your code through self-quizzing, transcript analysis, and validation flows. I'm Arjun, I'm currently a developer advocate at Pinecone, and I use agents to code a lot! I enjoy using agents to code, but I've been feeling like I've lost some conceptual learning and understanding that comes with writing the code yourself. I wanted to make a tool that would make it easy to re-build habits around learning, while complementing workflows that use agents. I found inspiration for the tool from reading a research paper on skill formation from Anthropic, and Shikhu is the result of my attempt to build a tool around that: https://arxiv.…

[dead]

I found verified, live credentials leaking from Microsoft, Google, Red Hat, Grafana, and LlamaIndex in public GitHub repositories.

LlamaIndex integration for SynapCores (RAG, GraphRAG, and hybrid retrieval)

We built a LlamaIndex integration for SynapCores, an AI-native database that combines SQL, vector search, graph traversal, and AI functions in a single engine. The integration supports RAG, GraphRAG, hybrid retrieval, metadata filtering, and SQL-based querying via a single backend. We'd appreciate feedback on the API, developer experience, and any features you'd like to see.

LlamaIndex integration for SynapCores (RAG, GraphRAG, and hybrid retrieval)

Ask HN: What are some rock solid open source vector databases

Hi timvisee, Thanks for your work on Qdrant! I actually don't really have any complaints at this time. I'm using it in conjunction with Qwen3-VL to autocaption/describe images and then place them in a large vector Qdrant db so its easy to semantically search through ~100k worth of image assets.

[dead]

Hi HN, I created AccuVideo because I couldn't find a privacy-respecting way to query and search inside my video library without uploading gigabytes of data to third-party clouds. AccuVideo runs on your local machine to index, transcribe, and let you chat with your video content. The indexed points are vectors are stored in Qdrant (cloud). We are launching today on Product Hunt and I'm offering free trials (generating licenses manually every hour). I would deeply appreciate the HN community's feedback on how to improve the local indexing pipeline. Also in product hunt today. Thanks for your time.

Mistral OCR 4

(jerry from llamaindex here) we're gonna benchmark on ParseBench and report the results!

Ask HN: What are some rock solid open source vector databases

Dev of Qdrant here. What would change that to 'very happy' in your use case?

Ask HN: What are some rock solid open source vector databases

I run a dockerized version of qdrant for RAG related stuff and have been pretty happy with it. https://github.com/qdrant/qdrant

Show HN: I built a free API cost calculator after a $340 surprise invoice

I was building a small LLM-powered side project and forgot to account for my system prompt tokens. 800 tokens × 2,000 requests/day × input token rate = $340 invoice I wasn't expecting. Built APICalculators to avoid doing this math in spreadsheets every time. It now covers 16 calculators: LLM APIs (GPT-4o, Claude, Gemini), vector databases (Pinecone, Qdrant, Supabase), serverless (Lambda, Cloudflare Workers), auth providers (Clerk, Auth0), payment processors (Stripe vs Paddle), and cloud VPS comparison. Everything runs in the browser — no signup, no data sent anywhere. Would love feedback on what's missing or wrong.

Ask HN: Who wants to be hired? (June 2026)

Location: Bucharest, Romania & San Francisco, CA Remote: Yes (preferred) Willing to relocate: Yes Technologies: Python, Django, Svelte / TypeScript, Shell, Linux, PostgreSQL, Redis, Docker, Hugging Face, OpenAI, LlamaIndex, LangChain, AWS, GCP, Unity (C#), Git, Github Actions, Software Design & Architecture, Devops, Product, UX, Sales Résumé/CV: Available on request Email: hi [at] lucianlazar [dot] io GitHub: https://github.com/xucian hey, I'm Lucian, have 13y of exp as a generalist swe, started as a gamedev then expanded into devops, ai/ml, trading, now working on a pretty diverse range of projects, from games, gamedev courses, ai tools, saas templates etc. more info on my profile looking for a fractional cto role, ideally, but I'm flexible

Show HN: AgentKitten: Swift package for provider-agnostic AI agents

If you’ve been building features with LLMs over the last few years you’ve likely used frameworks like LangChain/LangGraph, Vercel AI SDK, LlamaIndex, Google ADK, etc. Alternatively, you may be deeply exposed to chatbots, coding agents, and/or assistants. All of the above, roughly circling around the same idea, are flavors of the AI Agent loop being a mix of large language models and software scaffolding (typically called a harness). What’s surprising is that for Swift and Apple platform developers, a solid version of the harness-building toolset doesn’t seem to exist yet. Working on some exploratory features, I kept having to relearn and rebuild the same scaffolding every time I wanted to try a different provider. The kind of thing a framework exists to solve. And since a clean …

Ask HN: Who wants to be hired? (June 2026)

Location: Oslo, Norway (CET) Remote: Yes (EU/US time zones) Willing to relocate: Yes. Technologies: Rust, Python (PyTorch, NumPy/Pandas, Flask, Django), PyO3 Databases/infra: Postgres, ClickHouse, Qdrant, Redis, Kafka, Docker, Kubernetes (k8s) ML and systems: vLLM, TensorRT, Ray, edge ML inference, gRPC/Protobuf, systems profiling (perf, flamegraphs, Valgrind) Rust focus: Async services (Tokio, Axum/Actix), data/compute (Polars/Arrow), database proxies, performance-critical pipelines. Domain expertise: - Academic: PhD in mathematics (topological data analysis, representation theory), MSc finance, MSc neuroscience. - Applied: database proxies/networking, quant dev and backtesting engines, ML/DL for banking & finance (risk/fraud), LLMs and a…

Ask HN: Who wants to be hired? (June 2026)

Location: Munich, Germany Remote: Yes (open to remote, hybrid, or onsite) Willing to relocate: Yes — within Germany, or to Australia or USA Technologies: TypeScript, C#, Rust, Python, C++. Distributed systems and event-driven microservices — Kafka, RabbitMQ, MQTT, gRPC, GraphQL, Actor Model, CQRS, CRDTs. Data infra — TDengine, PostgreSQL, MongoDB, Redis, Elasticsearch, Qdrant, SIMD/AVX. .NET/ASP.NET Core, Node.js/NestJS, React/Next.js/React Native. Kubernetes, Docker, Terraform, Pulumi, AWS, Azure, GCP, CI/CD. Résumé/CV: https://shafiqahmad.com Email: ahmad2shafiq@icloud.com

Ask HN: Who wants to be hired? (June 2026)

Location: Europe (CET) Remote: Yes Relocation: No Technologies: Python, SQL, FastAPI, ETL, AI/LLM systems Résumé/CV: Available on request Email: inquire.hn@pm.me Backend operator with experience in enterprise systems, operations, and product execution. Most of my work has been in ambiguous, high responsibility environments requiring technical depth, judgment, systems thinking, and coordination across technical and business domains. MSc Quantitative Economics. Recently: • Built and deployed a production RAG platform (Python, FastAPI, React, Qdrant), including ingestion pipelines, hybrid retrieval, ranking/debugging workflows, and tooling to improve retrieval quality, reliability, and system stability as datasets scaled. Previously: • 10+ years building and owning backend, da…