Alex Reibman and Sri Laasya Nutheti 🖇️ showing people how to use the OpenAI Agents SDK and AgentOpsAI from Agency
Agency
Software Development
San Francisco, California 7,412 followers
Agency helps enterprises build, evaluate, and monitor AI agents. From the team at AgentOps.ai.
About us
Build reliable AI agents with monitoring, testing, and replay analytics. No more black boxes and prompt guessing. From the team behind AgentOps.ai
- Website
-
https://www.agen.cy/
External link for Agency
- Industry
- Software Development
- Company size
- 11-50 employees
- Headquarters
- San Francisco, California
- Type
- Privately Held
- Founded
- 2023
- Specialties
- ai, automation, B2B, data, dataentry, datamanipulation, datacleansing, datascience, datascientist, process, enterprise, and saas
Locations
-
Primary
Get directions
525 Market St
San Francisco, California 94105, US
Employees at Agency
Updates
-
Agency reposted this
Code > Concept: AGNTCY's Internet of Agents is now on GitHub Just weeks after launching AGNTCY, we're shipping code for the Internet of Agents, tackling the multi-agent software lifecycle from Discover—>Deploy: — DISCOVER: Open Agent Schema Framework and Agent Directory – to define agent capabilities and be able to catalog and search for agents at scale. — COMPOSE: Agent Connect Protocol, IO Mapper Agent and API Bridge Agent (as part of a multi-agent software toolkit) - to allow for agent-agent communication at the sematic and syntactic layers — DEPLOY: Agent Gateway and Workflow Server – for messaging and runtime on current cloud native state of the art — EVALUATE: in the works, but will tackle Observability, Evaluation, and Security frameworks especially for multi-agent systems. We are also welcoming 4 new members to the AGNTCY family: MongoDB, Traceloop, Agency, and Weaviate, alongside our current members Outshift by Cisco, LangChain, Galileo🔭, LlamaIndex, Glean Finally, we are introducing AGNTCY Supporter – a new role in the collective we’ve created for companies who believe in the mission of an open agent ecosystem. We're building the foundations that lets AI agents talk to each other regardless of who built them, where they run, or what framework they're using. Two big calls to action for you: 1. Read Papi Menon blog to understand how these components fit together and what they actually do 2. Try it out. Fork the repo, open issues, submit PRs. We need your participation to build an Internet of Agents that works for everyone, not just walled gardens. Blog & GitHub Links in comments 👇 #AGNTCY #InternetOfAgents #OpenSource
-
-
Agency reposted this
BIG AGNTCY news today = Code Release & New Community Members Today marks a significant milestone as the AGNTCY collective ships code just weeks after our initial announcement. Our code drop includes components in the agentic software lifecycle from Discover-->Deploy: For discovering agents with Open Agent Schema Framework and Agent Directory, composing them with Agent Connect Protocol, IO Mapper Agent, and API Bridge Agent as part of a multi-agent software toolkit, deploying at scale with Agent Gateway, and soon, evaluating their performance with integrated observability tools. We invite all developers to integrate these projects into their workflows today. We've built the "TCP/IP for AI agents." Plus, we welcome 4 new supporters—MongoDB, Traceloop, Weaviate, and Agency—organizations that believe in the open agent ecosystem mission. ➡️ Check out our blog for all the technical details and GitHub links: http://cs.co/6041FGcOv #AGNTCY #InternetOfAgents #OpenSource
-
-
Agency reposted this
Finally perfected the Engels language (A language made specifically for AI), although it holds tiny errors - The red is the original text - The blue is Engels - The green is saved with text within a context database for later This currently reduces text inputted by around x6 on average! Atop this, it is understood by almost any/all LLM systems! I created a language using OpenAI's GPT4 as the generation model, Agency's API for monitoring the training and Kaggle's journal system for running the training! As you can see, I inputted the conversation into a Character.AI session; besides one point which was incorrect (The amount of people who would be in the train), this system was able to properly remember the information though just the blue label! This will not be released yet, as it is directly made for Project Replicant (My Local AI System) but it will be FOSS eventually! Next up is turning this into a generation based onnx atop GPT2 (or similar) having the output fed to a tokenizer! Finally, i'll focus on running it locally with DeepSeek AI's R1 model or Qwen's 2.5 Omni!
-
-
Agency reposted this
Code > Concept: AGNTCY's Internet of Agents is now on GitHub Just weeks after launching AGNTCY, we're shipping code for the Internet of Agents, tackling the multi-agent software lifecycle from Discover—>Deploy: — DISCOVER: Open Agent Schema Framework and Agent Directory – to define agent capabilities and be able to catalog and search for agents at scale. — COMPOSE: Agent Connect Protocol, IO Mapper Agent and API Bridge Agent (as part of a multi-agent software toolkit) - to allow for agent-agent communication at the sematic and syntactic layers — DEPLOY: Agent Gateway and Workflow Server – for messaging and runtime on current cloud native state of the art — EVALUATE: in the works, but will tackle Observability, Evaluation, and Security frameworks especially for multi-agent systems. We are also welcoming 4 new members to the AGNTCY family: MongoDB, Traceloop, Agency, and Weaviate, alongside our current members Outshift by Cisco, LangChain, Galileo🔭, LlamaIndex, Glean Finally, we are introducing AGNTCY Supporter – a new role in the collective we’ve created for companies who believe in the mission of an open agent ecosystem. We're building the foundations that lets AI agents talk to each other regardless of who built them, where they run, or what framework they're using. Two big calls to action for you: 1. Read Papi Menon blog to understand how these components fit together and what they actually do 2. Try it out. Fork the repo, open issues, submit PRs. We need your participation to build an Internet of Agents that works for everyone, not just walled gardens. Blog & GitHub Links in comments 👇 #AGNTCY #InternetOfAgents #OpenSource
-
-
Agency reposted this
Hottest Startups of January 2025 🔥 Each month, we use the 'Hot Signals' feature within Specter to highlight companies and talent attracting the most interest from VC firms. 1. Bytespace.ai, founded in 2024 by Kar Dhillon and André Nuyens, is revolutionizing work automation with AI employees that can navigate and perform tasks on any website without APIs. Their mission is to empower solopreneurs and small teams to run complex operations efficiently, potentially transforming how businesses scale and operate in the digital landscape 🤖 2. Residua AI, founded in 2024 by Carmen Gutierrez and jon morales, is pioneering neural-enhanced knowledge management tools. With $1.7M in seed funding, they're transforming fragmented data into contextualized knowledge graphs, aiming to reduce cognitive overload in enterprise environments and reshape how organizations handle information 🧠 3. Retool, founded in 2017 by David Hsu, continues to dominate the low-code internal tool development space. With $190M raised and a $3.2B valuation, Retool is empowering developers to build custom applications 10x faster, serving 500,000+ customers including industry giants like Amazon and DoorDash 🛠️ 4. Find out more in the link below - Melty, Suno, Cognition, Scrapybara, Adam (YC W25), Perceptron AI, superwhisper, Nebesta, Open Vision Engineering, Agency, jazz.tools (Anselm Eickhoff), flocurve.com (Linus Ekenstam), beatsfoundation.com, Unblocked, Dawn, ell.so (William Guss), Santosh Sharan, Jacob Sansbury, Cosmin Nicolaescu, Firecrawl, and interaction.co (Felix M. Schlegel + Marvin von Hagen) #AIStartups #VentureCapital #TechInnovation https://lnkd.in/eK2yFpGY
-
Agency reposted this
Starting up a series of AI Hack nights in NYC is always a good idea, especially with great friends and partners like Flox Ramp Google Agency South Park Commons and Knock. If you're an engineer based in New York and want to join the next series of events, please be in touch!
-
-
Agency reposted this
Here is everything that happened in AI Agents this week 🧵 (save for later) 1/ Alex Reibman shared his vision for the modern AI Agent stack. https://lnkd.in/g78j8ipr 2/ “Interested to see how this year’s surge in AI agents, AI-driven creativity, and related innovations will shape—and potentially enhance— the market for things that are more "authentically human." - Reid Hoffman https://lnkd.in/grYX2QbC 3/ Big Agent release from Cohere coming this year. “We got something cooking :)” - Nick Frosst https://lnkd.in/g2Pwf-hm 4/ CopilotKit🪁 lets you turn any LangChain LangGraph agent into a user-interactive copilot. Great work Atai Barkai https://lnkd.in/gwAnuprD 5/ Sumanth P built a multi-agent AI news assistant using Open AI swarm and Llama 3.2. https://lnkd.in/gjpu_Gjc 6/ 2025 Will Be The Year of AI Agents! 👏 Bindu Reddy https://lnkd.in/g2ZP9C9i 7/ Wayne Hamadi shared a graphic about his views on AI agents. https://lnkd.in/grdFMXYk 8/ Qwen-Agent: Agent framework and applications built upon Qwen>=2.0. Thanks for sharing Ahsen Khaliq https://lnkd.in/gV2WGgpS 9/ A simple explanation the eliza framework works. https://lnkd.in/gWWs_Bx8 10/ “What if the AI SDK chat could do spreadsheets?" Nicolas Silberstein Camara https://lnkd.in/gMynZMca 11/ Sai Yashwanth shares his thoughts on which agentic framework works best for learning... https://lnkd.in/gt_jbYpY 12/ The team at Agency is bullish on Indian Agent devs in 2025. Happy to see Alex Reibman featured in the economic times. https://lnkd.in/gcKkhgXq 13/ Pydantic AI Agents Tutorial 🤖 https://lnkd.in/gwZCSfp6 14/ A Chrome extension to estimate calories directly on online menus. https://lnkd.in/gfwjRJjj 15/ Yasser Elsaid set a goal for Chatbase to reach $1M MRR by the end of the year... https://lnkd.in/gJ29VW6U 16/ Wayne Hamadi shared a guide for building useful vertical AI agents by putting humans in the loop https://lnkd.in/gHQfCmWd 17/ Mervin Praison ✅ released PraisonAI workflows https://lnkd.in/gf4napKy 18/ Shubham Saboo made two AI Agents play tic-tac-toe while the third agent watched as a judge. https://lnkd.in/gtTQgFUD 19/ Guohao Li introduces OASIS: Open Agent Social Interaction Simulations with One Million Agents. https://lnkd.in/gF8WHKg7 20/ Hesam Sheikh Hassani shares this 2025 AI engineer reading list: It includes 50 papers across 10 fields including AI Agents. https://lnkd.in/gMKG5JrW 21/ “Literally wish I could share the use cases we are seeing devs build AI Agents for.” - Adam Silverman (Hiring) https://lnkd.in/gpzFFygK 22/ Smolagents: lightweight library for building AI agents that generate and execute Python code. 🙌 Tom Dörr https://lnkd.in/gFPiu-C9 23/ Braelyn Boynton shared stats on the top AI Agent frameworks. https://lnkd.in/gU7DmY69 24/ Did I miss anything? Share below. Liked this thread? Share it with your friends/team. Follow Adam Silverman (Hiring) Silverman and Agency for everything AI agent related. Last weeks thread: https://lnkd.in/g4HAjace
-
-
Agency reposted this
Don't fall into the death zone of AI agents. You can sell an agent as a self-served product. You can provide high-ticket services to build agents in house. But don't fall in between. Don't sell yourself short when you build agents in-house. Do you know who will suffer the most from it? Your customers. Because if you've picked a critical activity to automate (which I hope you did, otherwise you will have no business impact), you're going to be involved A LOT as a service provider. And you're going to have to pick between getting more customers at the risk of decreasing the quality of your service, or increasing your price, which you should have done right from the beginning. Building agents in-house is very expensive in general, because it's about operationalizing and then automating a human activity. It can't be done without being constantly involved. The price point that separates the death zone from the good in-house agents projects is much higher than traditional software. So be careful, do not underestimate how much work is required. PS: comment agen(.)cy or DM me personally, and I will share with you what our customers tend to automate the most. Spoiler alert: your biggest gains might come from automating your 3rd most expensive activity instead of the first one, simply because it might be easier to implement.
-