Narendra

Building AI Agents with a mission to automate everything that i do !!

[Narendra @Narendra_SoniK]

Just got awarded with runner up at @eth_seoul_ in scalability track. It’s pleasure to build something useful for community. #web3 #blockchains #hackathon #KBW2022

[Narendra @Narendra_SoniK]

Whatever we do in life if we bring sincere effort and real passion to it, results follow. Sooner or later, they always do.
Grateful to be living this life, giving my best, and tasting the happiness that comes with it.

[Narendra @Narendra_SoniK]

True character is built by the values you refuse to compromise.
Discipline isn't about appearance, it's about decisions made when no one is watching.
Your habits become your identity. Your values become your legacy.

[Akshay 🚀 @akshay_pachaar]

the four types of agent loops.

loop engineering keeps getting talked about as one thing. it's actually a choice between four structures, and each one fits a different kind of task.

it means designing the system that steers the agent, instead of steering it yourself move by move.

that system always answers two questions. what starts a run, and what decides the work is done.

in a hand-run session you answer both yourself, every single time. each loop type moves more of that into the system.

here's each type, what triggers it, and when to reach for it.

1) turn-based.

triggered by a user prompt. the agent gathers context, acts, and checks its work inside a single turn, then a human reviews the output and writes the next prompt.

use this when requirements are still forming and every output changes what you'd ask for next.

2) goal-based.

triggered by a /goal command carrying success criteria and a budget, like "get the homepage Lighthouse score to 90, stop after 5 tries." when the agent tries to stop, an evaluator model checks whether the goal is met, and a no sends it back to work.

use this when the outcome is measurable but the path there isn't worth your attention.

3) time-based.

triggered by a clock. an interval fires, the agent runs a fixed prompt like "check the PR, fix CI," then waits for the next tick. /loop runs on your machine, /schedule moves it to the cloud so it survives a closed laptop.

use this for recurring work where the task is known in advance and only the timing repeats.

4) proactive.

triggered by an event or schedule with no human present. a routine watches a channel, and when something needs handling it spawns a workflow with a triage agent, a fix agent, and a reviewer that adversarially judges the work before the task closes.

use this for standing responsibilities where you can't predict what will come in, only that something will.

each type hands off one more job than the last. turn-based keeps both with the human, goal-based automates the checking, time-based automates the trigger, and proactive automates both while deciding the workflow shape at runtime.

so the mapping question isn't which loop is most advanced. it's whether your task is exploratory, measurable, recurring, or standing.

the more you hand off, the less you babysit.

I wrote the full breakdown on loop engineering. the article is quoted below.

[Narendra @Narendra_SoniK]

If you’re in Delhi-NCR, I would love to connect with you to build tech community, run meetups, tech sharing sessions.
I build #ai agents, data pipelines, scalable SaaS, data intensive applications. Looking forward to connect with like minded folks here.
#Community #tech #india

[Narendra @Narendra_SoniK]

Looking forward to connect with Data Scientists, data engineers to join us in building AI Agents at @zimeai

#hiring

[Narendra @Narendra_SoniK]

While everyone is talking about AI Agents and LLMs. There are so many interesting challenges to solve in search, recommendations and analytics.

At @zimeai we are hiring for both #search and #ai roles.

Please send applications at [email protected]

#hiring #salestech

[Narendra @Narendra_SoniK]

I am writing python scripts to clean noise from audio(.wav) files. I would love to connect with you if you have experience in Audio Engineering.

Let's connect.

#python #audioengineering #DL

[Narendra @Narendra_SoniK]

@kubernetesio I can’t imagine without k8s how i will manage my workloads efficiently

[Narendra @Narendra_SoniK]

If you would like to use the blueprint in @HashiCorp terraform, just DM.

[Narendra @Narendra_SoniK]

After building so many projects, I have a blueprint to build tech for any startup. My cloud infras gets provisioned from scratch while I take a coffee break.
It can scale from POC > MVP > Prod by just changing config and applying my Terraform IAC.

More details in thread below.

[Narendra @Narendra_SoniK]

@awscloud @kubernetesio It also auto creates GitHub actions CI/CD pipelines to build, push docker images to repository, promote new image to respective environments(dev/staging/prod). For front end apps it creates the assets and pushes it to S3 Bucket and invalidates cloudfront.

[Narendra @Narendra_SoniK]

Explaining a topic helps in learning it better and brings more clarity. If a new question comes while explaining then it accelerates the learning even further.

[Narendra @Narendra_SoniK]

Getting the "job done" and "doing it well" are two different perspectives that often come from an individual's personality, character, and attitude towards life in general.
Many can get the job done, but very few can do it well.