agents will help empower smaller startups to do more EAs are a great first step toward this "agent workforce" whats the next role to be made agentic?
@hwchase17
-
Mature Open Source Technology Adopted by Major Tech Companies
By
–
yes its pretty mature – used by replit, uber, linkedin, etc
-
FSM Mental Models for Agent Development and State Management
By
–
#1: i graphs/fsm is useful mental model/devX for building agents (this is a matter of taste though) #2: state management (short memory: easy, long term memory: medium, checkpointing: hard, human-in-the-loop: hard)
-
LangSmith for LLM Pipeline Cost Monitoring and Optimization
By
–
we see a lot of usage of langsmith for monitoring (and then reducing) costs of you LLM pipelines/agents
-
AI Email Assistant Ambient Agent Open Sourced
By
–
For the past six months, I haven't checked email directly, but rather have relied on an AI email assistant to triage and draft emails for me This is an example of an "ambient agent" We're open sourcing the code today, as well as a hosted version for people to try out
-
Ambient Agents: New Open Source Email Assistant Technology
By
–
Read more about ambient agents here: https://
blog.langchain.dev/introducing-am
bient-agents/
… See the open source email assistant here: https://
github.com/langchain-ai/e
xecutive-ai-assistant
… Use the hosted email assistant for free: https://
agentinbox.ai -
LangChain Developer Hiring Guide Released
By
–
Cool to see this!
— Harrison Chase (@hwchase17) 10 janvier 2025
Have had a few people ask me the best way to hire LangChain devs – now I know where to point them https://t.co/CJ3C0G5u9pCool to see this! Have had a few people ask me the best way to hire LangChain devs – now I know where to point them
-
Fleet Learning: Understanding Collaborative AI Model Training
By
–
'fleet learning' is cool term, havent heard tht before
-
Dynamic Few-Shot Prompting Improves Agent Performance
By
–
The promise of dynamic few-shot prompting After sharing AppFolio's story of putting an agent in production, @jobergum noted a key part of the story: how dynamic few shot prompting greatly improved their performance This is how it works: You collect a set of example inputs