Best Programming Language for Coding with AI Agents: Insights for Modern Development

AI agents are transforming software development, allowing developers to focus on high-level architecture while handling syntax and boilerplate code. At HariKrishna IT Solutions, we integrate AI-assisted coding into our .NET, SQL Server, and legacy modernization services to deliver scalable, cost-effective solutions for global clients.

This blog adapts a key discussion from Peter Steinberger, creator of the rapidly growing OpenClaw AI agent framework, with Lex Fridman. Presented as an extended transcript with analysis, it explores optimal languages for agentic programming—perfect for teams evaluating offshore outsourcing.

The Shift from Specialists to Builders

Peter Steinberger challenges traditional roles: “Like part of what I did this year is I went to a lot of iOS conferences because that’s my background and just told people don’t consider – don’t see yourself as an iOS engineer anymore. Like you need to change your mindset. You’re a builder and you can take a lot of the knowledge how to build software into new domains and all of the more fine details agents can help. You don’t have to know how to splice an array or what the—what the correct template syntax is or whatever, but you can use all your general knowledge and that makes it much easier to move from one galaxy, one tech galaxy into another.”

This mindset shift is crucial for offshore teams. Builders leverage AI to pivot across stacks, reducing onboarding time and costs. In our projects, this means a .NET expert can prototype Python ML services or Rust microservices seamlessly.

Language Recommendations for AI Agents

Steinberger selects languages based on ecosystem, performance, and agent compatibility, not personal preference. Here’s an expanded view:

“And often times there’s languages that make more or less sense depending on what you build, right? So for example, when I build simple CLI, I like Go. I actually don’t like Go. I don’t like the syntax of Go. I didn’t even consider the language, but the ecosystem is great. It works great with agents. It is garbage collected. It’s not the highest performing one, but it’s very fast. And for those type of CLIs that I build, Go is a really good choice.”

Lex responds: “Isn’t that fascinating that here’s a programming language you would have never used if you had to write from scratch and now you’re using because LMs are good at generating it and it has some of the characteristics that makes it resilient like garbage collected because everything is weird in this new world and that just makes the most sense.”

Full Extended Transcript: The Conversation Unfolds

Peter: What’s the best—ridiculous question? What’s the best programming language for the AI agentic world? Is it JavaScript/TypeScript? TypeScript is really good. Sometimes the types can get really confusing and the ecosystem is a jungle. So for web stuff, it’s good. I wouldn’t build everything in it.

Lex: Don’t you think we’re moving there? Like that everything will eventually be written, eventually it’s written in JavaScript. Births and deaths of JavaScript and we’re living through it in real time. Like what does programming look like in 20 years, right? In 30 years, in 40 years, what do programs and apps look like?

Peter: You can even ask a question like, do we need a programming language that’s made for agents? Because all of those languages are made for humans. So, what would that look like? I think there’s a whole bunch of interesting questions that we’ll discover and also how because everything is now world knowledge, how it in many ways things will stagnate ’cause if you build something new and the agent has no idea, that’s going to be much harder to use than something that’s already there.

Peter (on Swift): When I build Mac apps, I build them in Swift and SwiftUI, partly because I like pain. Partly because the deepest level of system integration I can only get through there. And you clearly feel a difference if you click on an Electron app and it loads a web view in the menu. It’s just not the same.

Peter (on experimentation): Sometimes I just also try new languages just to like get a feel for them like Zig. Yeah. If it’s something that where I care about performance a lot and it’s a really interesting language and agents got so much better over the last 6 months from not really good to totally valid choice. Just still a very young ecosystem and most of the time you actually care about ecosystem, right?

Peter (on Python and deployment): So, if you build something that does inference or goes into whole running model direction, Python—very good. But then if I build stuff in Python and I want a story where I can also deploy it on Windows, not a good choice. Sometimes I found projects that kind of did 90% of what I wanted but were in Python and I wanted an easy Windows story. Okay, just rewrite it in Go.

Peter (on Rust): But then if you go towards multiple threads and want more performance, Rust is a really good choice.

Peter (closing): There’s just no single answer and it’s also the beauty of it. Like it’s fun and now it doesn’t matter anymore. You can just literally pick the language that has the most fitting characteristics and ecosystem for your problem domain. And yeah, it might be you might be a little bit slow in reading the code, but not really. I think you pick stuff up really fast and you can always ask your agent.

Practical Applications in Enterprise Development

This discussion aligns perfectly with HariKrishna IT Solutions’ approach. In legacy modernization, we use AI agents to migrate VB.NET monoliths to Rust-backed microservices, preserving business logic while boosting performance. For SQL Server optimization, Python agents analyze queries, then Go tools automate indexing.

Consider a real-world scenario: A government client with aging ASP.NET apps. We prototype SwiftUI dashboards for macOS admins, TypeScript for web portals, and Rust for backend scalability—all agent-assisted. This hybrid stack cuts development time by 40%, enabling 30-70% cost savings through Indian offshore expertise.

Code example: Agent-generated Go CLI for .NET deployment checks.

package main

import (
    "fmt"
    "os/exec"
)

func checkDotNetVersion() string {
    cmd := exec.Command("dotnet", "--version")
    output, err := cmd.Output()
    if err != nil {
        return "Not installed"
    }
    return string(output)
}

func main() {
    version := checkDotNetVersion()
    fmt.Printf("Current .NET version: %s\n", version)
}

This simple tool verifies environments before deployments, scalable for enterprise CI/CD.

Future of Agentic Programming and Outsourcing

As Steinberger notes, stagnation risks exist if agents lack novel knowledge, but established ecosystems thrive. Offshore firms like ours bridge this by combining human oversight with AI speed. Expect languages like Zig to mature, Python for AI dominance, and Rust for perf-critical apps.

For e-commerce clients, TypeScript agents build scalable platforms; media firms get Swift-native tools. Our SQL Server migrations use Python for analysis, Go for tools, ensuring seamless transitions.

Why Choose HariKrishna for AI-Driven Development

Our team excels in full-stack .NET, database design, and modernization, now supercharged by AI agents. Achieve ROI through lower costs, faster delivery, and quality assurance.

Schedule a consultation to discuss your .NET outsourcing needs, legacy migration, or AI integration. Receive a customized proposal tailored to your ROI goals.