Akshay Deo (@akshay_deo) 's Twitter Profile
Akshay Deo

@akshay_deo

CTO @getmaximai l Ex @getpostman @slackhq @medianetads | Co-Founded @appsurfer (bit.ly/adeoas) @BetaCraft_Tech (acqui-hired)

ID: 63389707

linkhttps://www.akshaydeo.com?ref=twtr_profile calendar_today06-08-2009 06:07:21

3,3K Tweet

1,1K Followers

415 Following

Maxim AI (@getmaximai) 's Twitter Profile Photo

📣 📷 AI Evals Mixer | May 21st Maxim AI is excited to host the AI Evals Mixer - Bangalore edition in association with Razorpay Ventures and Mars Shot VC on May 21st, bringing together engineering and product leaders shaping the future of AI. Join us for an evening of

📣 📷 AI Evals Mixer | May 21st

<a href="/getmaximai/">Maxim AI</a> is excited to host the AI Evals Mixer - Bangalore edition in association with <a href="/Razorpay/">Razorpay</a> Ventures and <a href="/MarsShotVC/">Mars Shot VC</a> on May 21st, bringing together engineering and product leaders shaping the future of AI.

Join us for an evening of
Maxim AI (@getmaximai) 's Twitter Profile Photo

With Maxim’s one-line integration for CrewAI , you can easily trace and analyse agent interactions, performance metrics, and more. Features: One Line Integration 1. End-to-End Agent Tracing: Monitor the complete lifecycle of your agents 2. Performance Analytics: Track

With Maxim’s one-line integration for <a href="/crewAIInc/">CrewAI</a> , you can easily trace and analyse agent interactions, performance metrics, and more. 

Features: One Line Integration

1. End-to-End Agent Tracing: Monitor the complete lifecycle of your agents
2. Performance Analytics: Track
Maxim AI (@getmaximai) 's Twitter Profile Photo

This comprehensive guide will walk you through creating an intelligent event-discovery agent (an agent that discovers public events happening in the US) using n8n.io (an open-source workflow automation platform) and rigorously testing it with Maxim AI (an agent testing

This comprehensive guide will walk you through creating an intelligent event-discovery agent (an agent that discovers public events happening in the US) using <a href="/n8n_io/">n8n.io</a>  (an open-source workflow automation platform) and rigorously testing it with <a href="/getmaximai/">Maxim AI</a>  (an agent testing
Maxim AI (@getmaximai) 's Twitter Profile Photo

Let's build an AI Agent ❌ ..... a well tested AI Agent ✅ In this video we will be creating an intelligent event-discovery agent (an agent that discovers public events happening in the US) using n8n.io (an open-source workflow automation platform) and testing it with

Maxim AI (@getmaximai) 's Twitter Profile Photo

Do Language Models Know That They're Being Evaluated? Picture this scenario: You’re very new to AI, exploring chatgpt by testing its capabilities on various topics, expecting honest answers unaware that behind the scenes, it already figured out that it’s being tested and is

Do Language Models Know That They're Being Evaluated?

Picture this scenario: You’re very new to AI, exploring chatgpt by testing its capabilities on various topics, expecting honest answers unaware that behind the scenes, it already figured out that it’s being tested and is
Maxim AI (@getmaximai) 's Twitter Profile Photo

Last week at Maxim AI, we rolled out several powerful upgrades to give teams more control, clarity, and customization across the platform. Here's what’s new: Custom Dashboards Just Got an Upgrade ⬆️ Dashboards are now more flexible and insightful: - Custom metric cards

Last week at <a href="/getmaximai/">Maxim AI</a>, we rolled out several powerful upgrades to give teams more control, clarity, and customization across the platform. 

Here's what’s new:  

Custom Dashboards Just Got an Upgrade ⬆️
Dashboards are now more flexible and insightful: 
- Custom metric cards
Maxim AI (@getmaximai) 's Twitter Profile Photo

Maxim's Integration for Anthropic is Here! 🚀 Build AI systems using Anthropic Claude LLMs and effortlessly push logs to Maxim AI. Monitor the following data about your agent run on Maxim Dashboard - - Costs - Latency - Token Usage - LLM activity - Function Calls

Maxim's Integration for <a href="/AnthropicAI/">Anthropic</a> is Here! 🚀

Build AI systems using Anthropic Claude LLMs and effortlessly push logs to <a href="/getmaximai/">Maxim AI</a>. 

Monitor the following data about your agent run on Maxim Dashboard - 
- Costs
- Latency 
- Token Usage
- LLM activity 
- Function Calls
Akshay Deo (@akshay_deo) 's Twitter Profile Photo

Most agent observability setups today are missing a critical piece: they trace LLM calls in isolation, completely disconnected from the broader API or system traces. The result? Garbage in, garbage out. Agents aren't just a chain of LLM calls; they’re cohesive systems

Akshay Deo (@akshay_deo) 's Twitter Profile Photo

A public company recently deployed Bifrost to their staging environment swiftly and without any support from our team. The result? Over 50x improvement in latency. Companies value flexibility and speed, while developers prioritize quality and efficiency. Bifrost delivers on all

A public company recently deployed Bifrost to their staging environment swiftly and without any support from our team.

The result? Over 50x improvement in latency.

Companies value flexibility and speed, while developers prioritize quality and efficiency. Bifrost delivers on all
Maxim AI (@getmaximai) 's Twitter Profile Photo

Remember when AI models just autocompleted your emails and churned out some Python code blocks? Those were simpler times. Now we're living in the golden age of AI agents where Cursor makes developers feel superhuman, Harvey prints enterprise money raising to $5B

Remember when AI models just autocompleted your emails and churned out some Python code blocks? Those were simpler times. Now we're living in the golden age of AI agents where <a href="/cursor_ai/">Cursor</a>  makes developers feel superhuman, <a href="/harvey__ai/">Harvey</a> prints enterprise money raising to $5B
Akshay Deo (@akshay_deo) 's Twitter Profile Photo

Getting tool call accuracy right is key for a smooth Agent UX. In our latest benchmarking post (link in the comments), we break down how adding more context or tools to your prompts can actually make accuracy drop from 73 percent to 66 percent. Want to keep your agents sharp?