Full-Stack Developer

Neohumans.ai

Overview

Neohumans was my first true individual contributor role.

Unlike previous roles where the challenge was learning software engineering or understanding business operations, this role was about adapting quickly, shipping fast, and solving whatever problem the product needed solved next.

The company moved through multiple product explorations during my time there.

Each new direction introduced unfamiliar challenges, technologies, and constraints.

Rather than specializing in a single area, I found myself moving across frontend development, backend systems, analytics, AI integrations, deployment pipelines, real-time communication, and eventually hardware-adjacent experimentation.

Looking back, the defining lesson of this period was adaptability.

When the product changed, the engineering challenges changed.

The job was to learn quickly and keep moving.

The Journey

Shipping Fast

One of my earliest experiences at Neohumans was helping prepare a beta release shortly after joining the company.

I was still unfamiliar with the codebase, yet the expectation was clear.

Understand the system.

Find what matters.

Ship.

That set the tone for the rest of the role.

AI Chat Products

A major focus became building AI-enabled chat experiences.

I worked on frontend systems, backend integrations, analytics instrumentation, real-time communication workflows, and OpenAI-powered interactions.

The work required balancing user experience, performance, reliability, and product experimentation.

This period also exposed me to operating systems at significantly larger scale than anything I had previously worked on.

The platform served millions of users and generated millions of analytics events every month.

Performance & Scale

As the platform grew, performance became increasingly important.

One notable improvement involved reducing the frontend build size from approximately 60 KB to 12 KB.

The experience reinforced an important lesson:

Performance is not just an engineering metric.

At scale, performance directly affects user experience.

Analytics & Product Learning

Implemented Amplitude instrumentation across the platform.

The system captured approximately 6 million events per month.

This provided visibility into how users actually interacted with the product.

It strengthened my appreciation for data-informed product development.

Build
→ Measure
→ Learn
→ Improve

Real-Time Systems

Built WebSocket infrastructure in Go to support real-time communication workflows.

This expanded my responsibilities beyond frontend development and introduced practical challenges around connection management, backend communication patterns, streaming experiences, and latency-sensitive systems.

I also worked on OpenAI streaming integrations, helping bridge traditional software systems with AI-driven experiences.

Product Pivots & Experimentation

One of the most interesting parts of Neohumans was experiencing how quickly product direction can change.

The company explored multiple opportunities, each with different technical requirements.

Rather than viewing these changes as disruptions, they became opportunities to learn new domains.

Every pivot required understanding a new problem space and finding practical ways to contribute.

AI Robotics & Conversational Toys

The most unusual challenge emerged when the company began exploring AI-powered robotics and conversational toys.

My role evolved into building software layers that connected hardware systems with backend services.

This included experimentation with ESP32 devices, Raspberry Pi systems, audio pipelines, real-time communication, and conversational interfaces.

The challenge was not simply making hardware work.

The challenge was creating a bridge between physical devices and intelligent software systems.

As technical and operational constraints became clearer, the product direction evolved toward simpler and more practical approaches.

That process taught me a valuable lesson:

Good engineering is not about building the most complex solution.

It is about finding the solution that best fits reality.

Key Impact

• Helped ship a beta product shortly after joining.
• Built and scaled AI-enabled chat experiences.
• Supported a platform serving approximately 20 million users.
• Reduced frontend build size from approximately 60 KB to 12 KB.
• Implemented analytics instrumentation capturing approximately 6 million events per month.
• Built WebSocket infrastructure for real-time communication.
• Integrated OpenAI streaming APIs.
• Worked on AI robotics and conversational toy initiatives.
• Connected hardware systems with backend software services.
• Contributed across frontend, backend, infrastructure, analytics, AI, and hardware domains.

What This Role Taught Me

Neohumans taught me how to operate in uncertainty.

Products evolve.

Business priorities change.

New technologies emerge.

Engineering teams are constantly asked to solve problems they have never seen before.

The most valuable skill is not knowing everything in advance.

The most valuable skill is being able to learn quickly and become useful in unfamiliar situations.

This role also deepened my understanding of scale, product experimentation, AI systems, real-time communication, and cross-domain engineering.

Most importantly, it reinforced a principle that continues to guide my work:

When the problem changes, adapt.

When the system needs something unfamiliar, learn it.

When the product evolves, evolve with it.

Why This Experience Matters

Neohumans represents the phase where I moved beyond being primarily a frontend engineer and became a more adaptable product engineer.

It combined AI, scale, experimentation, infrastructure, real-time systems, and hardware exploration into a single experience.

More than any specific technology, it taught me how to learn new domains quickly and contribute wherever the product needed me most.