Founding Engineer

Founding Engineer

Be our founding engineer in India: build the core AI system end to end, own the architecture, and set the technical bar—an individual-contributor role with founding equity, working directly with the CEO. If you want your decisions to shape a company (not just a codebase) and you care about reliable AI systems that hold up in production, this is your seat.

Blaugarnet is building the Intent Layer for AI-driven software—capturing business intent from messy inputs, running it through a governed, human-in-the-loop deliberation process, and producing human-authorized blueprints an AI agent can build from correctly, with a permanent append-only record of decisions. Our enterprise product is live in early access; a self-serve product is on the near roadmap. We’re small, early, and moving fast—and this hire is part builder, part architect, part technical anchor.

The Work

• Build the core AI system end to end: the reasoning and deliberation engine plus the context engineering around it (planning, tool use, memory, validation).
• Own the retrieval and data pipelines: retrieval over structured and unstructured data, designed to support correctness and traceability—not just “it works on my machine.”
• Take the system from demo-grade to production-grade: scalable, observable, cost-aware AI services that enterprise customers can depend on.
• Set the architecture and technical direction: choose frameworks, patterns, and infrastructure—including the system-of-record data architecture and the on-prem/VPC/regulated deployment posture—and live with the consequences.
• Raise the technical bar as the standard-bearer: architecture and code quality reviews, reliability focus, and engineering practices the India team will inherit.

What You’ll Need

• Deep backend engineering experience in a modern typed/async stack (we use Python/FastAPI today), with strong service and API design.
• AI-native development habits—you work with coding agents (Claude Code, Cursor, or similar) to design, build, and verify production systems.
• Production experience building GenAI systems that real users rely on (not prototypes).
• Deterministic orchestration of LLM workflows: control flow, authority, and guardrails engineered in machinery (state-machine/workflow-engine style), not left to the model’s discretion.
• Governance enforced outside the model: deterministic checks that keep the system’s rules intact even when the model output is messy.
• RAG in production: vector databases (e.g., pgvector), embedding pipelines, and retrieval over mixed data types.
• Relational data modeling under correctness constraints: systems of record with append-only/immutable patterns, provenance/lineage integrity, and strong SQL/Postgres.
• Evaluation-harness design: eval sets/benchmarks/regression checks that prevent silent quality degradation when models or prompts change.
• Cost and performance discipline: token tracking, model-selection heuristics, caching, and practical trade-offs.
• Zero-to-one ownership: you’ve defined the spec and made architecture calls under real uncertainty.
• Technical leadership as an IC: you raise the bar through reviews, direction, and mentorship—without needing a manager title.

Even Better If

• Experience with MCP and agent interoperability standards.
• You’ve built in regulated or high-stakes enterprise domains (governance, audit, compliance).

Who Thrives Here

• You’re comfortable being the person who turns ambiguity into an architecture—and you keep ownership until the system is actually working reliably.
• You guide others in a supportive, non-directive way: you clarify options, provide the tools and context, and enable independent good decisions.
• You communicate with clarity and directness, and you build trust by checking in to make sure the team has what it needs to move.
• You think in structured, region-by-region execution—reviewing the system carefully, then producing concise, actionable outcomes.

The Team

You’ll be the technical anchor for the India build, working closely with the CEO and shipping the core of the product from the ground up. Early on, you’ll be deep in code every day; over time, you’ll increasingly shape engineering practice and technical direction for the team. This is an individual-contributor role—high ownership and influence, without people-management responsibilities.

Comp, Logistics and Benefits

Base + bonus target ₹50–₹60 LPA + meaningful founding equity (additional, separately listed). Full-time. Pune, India (hybrid)—the team can relax to fully remote initially only if they can’t shortlist enough qualified candidates for the Pune hybrid role.

• Founding equity (meaningful; additional to base/bonus)
• Performance bonuses
• Health, dental, and vision benefits
• Paid time off
• Professional development support

Non-Traditional Backgrounds

If you’ve built and scaled AI-backed systems in any environment—startup, enterprise, or open source—and you can show how you engineered reliability, governance, and correctness, we want to hear from you.