Founding Engineer (Full Stack / Systems)

About Jnaara

We're not building another AI app. We're building an AI-native research system that emulates how top investors think — transforming complex data, ideas, and workflows into structured, decision-grade outputs. This is a systems + infrastructure problem, not a wrapper.

Jnaara is built by a team of veteran researchers, portfolio managers, and CTOs from renowned hedge funds and asset management firms. We're working closely with a $200B+ global asset management firm as a co-build partner — designing for real workflows, real constraints, and real users from day one. You won't be assembling tools — you'll be defining how an entire research system thinks and operates. We're building systems where correctness, auditability, and reasoning quality matter — not just UX.

The Technical Challenge

You'll be working on a platform where XXX+ AI agents today (rapidly scaling) collaborate across complex, multi-step workflows — each with different tools, data access patterns, and reasoning strategies.

• Workflows are long-running, stateful, and non-deterministic
• Outputs must be reproducible, explainable, and auditable
• Systems must balance latency, cost, and reasoning quality

This is not prompt chaining. This is orchestrating intelligent systems under real-world constraints.

What You'll Build

• Design systems coordinating interacting agents across dependency graphs, retries, and evaluation loops
• Build abstractions for workflows (not chat chains) — inter-agent communication, tool delegation, and error recovery
• Implement context and memory systems: state persistence, retrieval layers, and reasoning traces
• Architect scalable pipelines that transform complex, heterogeneous data into structured outputs
• Design flexible data access layers for dynamic, agent-driven analysis
• Enable large-scale experimentation with reproducibility and performance in mind
• Build async-first backend services (Python / FastAPI) handling concurrent workflows, long-running jobs, and high-throughput processing
• Design task orchestration, caching (Redis), queuing (Celery), and compute pipelines
• Architect for bursty workloads and hybrid compute (batch + real-time)
• Implement tracing, latency profiling, and usage monitoring
• Build evaluation pipelines for output quality and system performance
• Make AI systems debuggable, inspectable, and auditable at every layer
• Build real-time, data-rich interfaces (React / Next.js) for interacting with complex workflows
• Design UX for inspecting intermediate outputs, comparing results, and configuring systems
• Stream intermediate results (WebSockets / SSE) as workflows execute
• Own the design system and component architecture
• Own cloud infrastructure (AWS) — compute, storage, networking, and security
• Build CI/CD pipelines, automated testing, and deployment workflows
• Implement infrastructure-as-code for reproducible environments
• Design for data governance: encryption, RBAC, audit logging

Tech Stack (Current Direction)

• Backend: Python, FastAPI, Celery, Redis
• Frontend: React, Next.js, TypeScript
• Data: Snowflake, Postgres, S3
• AI Layer: Multi-agent orchestration, retrieval systems, LLM APIs
• Infra: AWS, Terraform, GitHub Actions

What We're Looking For

• 6–8 years building production-grade systems
• Strong in Python (APIs, async systems, data workflows) and React / Next.js
• Thinks in systems, not endpoints
• Comfortable across backend, data, and frontend layers
• Has built something from 0 1
• Hands-on with cloud infrastructure and modern DevOps
• Strong data instincts (SQL, modeling, performance)
• High ownership, fast iteration mindset

Strong Signals

• Experience working with LLMs or AI systems in production
• Familiarity with data pipelines or async job orchestration
• Real-time systems or event-driven architecture experience
• Startup or founding engineer experience
• Interest in complex decision-making systems or research workflows

Technical Problems You'll Tackle

• Orchestrating non-trivial multi-agent systems with real interdependencies
• Designing memory and context layers for reasoning systems
• Balancing latency vs cost vs quality in AI workflows
• Making outputs traceable, reproducible, and debuggable
• Building systems where correctness matters as much as speed

Compensation

• 35–60 LPA + meaningful founding equity (0.25–1.5%)
• Full ownership of core systems and architecture
• Direct exposure to real users solving high-stakes problems from day one

Why This Is Different

• Most AI startups: wrap APIs, optimize prompts, ship demos
• We're building: a research engine with real institutional users solving high-stakes problems where systems thinking > prompt engineering

If you care about building systems that think, not just respond, we should talk.