Reply by: StartupFounder_Alex
We built custom solution using LangChain + GPT-4 + custom training on our data. Not plug-and-play unfortunately.
For off-the-shelf solutions, look at:

Adept AI
AutoGPT
AgentGPT
BabyAGI

All still early stage tho. This tech probably 1-2 years away from mainstream adoption but moving fast.

Reply by: FutureTechFan
Valid concerns from both sides. @EthicalTechAdvocate I hear you on job displacement - its real issue we need to address as society.
But also cant stop progress. Better question is how do we manage transition so people arent left behind?
@StartupFounder_Alex - can you share which tools/platforms your using for agentic AI? Im interested to test for our business.

Reply by: EthicalTechAdvocate
This is exactly what im worried about. 80% automation means 80% of customer support jobs disappear. Yes its efficient, yes it saves money, but what about the people?
Also, if AI is making decisions that affect customers and it screws up 10-15% of time, that's alot of unhappy customers. Human error is one thing, systemic AI error affecting everyone is different problem.
We need to think about implications beyond just "its cool and saves time"

Reply by: StartupFounder_Alex
Already using agentic AI in our operations and its been game-changer. We use it for:

Customer support (AI agent handles 80% of inquiries without escalation)
Code review (AI agent reviews PRs, suggests improvements, even writes tests)
Data analysis (Give it business question, it queries databases, analyzes, creates reports)

Not perfect - still makes mistakes maybe 10-15% of time. But improves every month and saves huge amount of time.
The key is proper guardrails. Don't give it unrestricted access to everything. Set boundaries, monitor closely, have human review for critical decisions.

Reply by: PracticalEngineer_Tom
Sounds cool but also scary. AI making decisions independently without human oversight? What could go wrong lol.
Seriously tho, who's liable when agentic AI makes wrong decision? The company? The AI developer? The user? Legal framework not ready for this.

Reply by: AIResearcher_Nina
Agentic AI is real and coming fast. Microsoft CEO recently talked about how AI agents will transform how we work. Agentic AI has transformative potential to redefine testing paradigms, and same applies to other domains.
Difference from current AI:

Current AI: Responds to specific commands/prompts
Agentic AI: Takes high-level goal and figures out steps independently, learns continuously, connects across systems

Example: Instead of telling AI "run this test on this browser", you tell it "make sure our app works on all platforms" and it figures out what tests to run, when to run them, how to fix issues it finds.

Started by: FutureTechFan | Jan 2025
Keep seeing term "Agentic AI" everywhere. From what I understand, its AI that can work independently, make decisions, and complete complex tasks without constant human supervision.
Is this actually happening now or still experimental? How is it different from current AI automation?