Senior Backend Engineer

PKC Management Consulting; Chennai District, Bengaluru; On-site; INR 4,000,000 - 7,000,000; Full time

Job summary

Piaxis is hiring a Senior Backend Engineer to own core services in our agent platform. You will ship production-grade Python services (FastAPI) and shape multi-step pipelines with LangGraph, PostgreSQL, and Weaviate, collaborating closely with founders.

You will operate near the metal of an LLM-driven system, focusing on retrieval, evaluation harnesses, and scalable architecture, with ADRs guiding decisions and a strong emphasis on metrics over vibes.

Qualifications

• 5+ years building production backend systems with strong Python depth.
• Experience designing API services and asynchronous/multi-step pipelines (FastAPI or equivalent).
• Solid grasp of data modeling with relational DBs (PostgreSQL) and vector/search systems.
• Track record owning systems end to end: design, ship, measure, iterate.
• Evaluation-driven mindset: build test harnesses and metrics to prove improvements.

Responsibilities

• Build and own backend services in Python with FastAPI and LangGraph pipelines.
• Design and strengthen the agent pipeline using vector search (Weaviate) and embeddings.
• Extend our eval framework to measure improvements rigorously.
• Integrate multiple LLM providers (Anthropic, OpenAI, Bedrock, Google) with LangGraph.
• Shape architecture, write ADRs, and keep systems extensible and maintainable.

Skills

Python development

FastAPI

API design

Distributed systems

Evaluations/metrics

Tools

PostgreSQL

Weaviate

LangGraph

LangChain

Docker

AWS

Job description

About Piaxis

Piaxis builds AI agents for the architecture, engineering, and construction (AEC) industry. We're solving a $150B problem in US construction documentation that remains largely untouched by software. Our Detail Library product is already live with paying customers, and our AI Detailing Agent is on track to launch - replacing weeks of manual drafting work with minutes.Founding team: an ex-McKinsey operator, a US-licensed architect (ex-Perkins Eastman), and ex-technical cofounder of Belong.co - domain depth plus technical depth to build this right.

The Role

We're hiring a Senior Backend Engineer (5+ years) to own core services in our agent platform - the pipelines that orchestrate LLM reasoning, retrieve the right reference details, place elements accurately, and prove it with evals. You'll work close to the metal of an LLM-driven system: retrieval quality, pipeline architecture, data contracts, and the evaluation harness that tells us whether the agent is actually getting better. This is a high-ownership IC role. You'll ship production services, shape architecture decisions (we write ADRs before code lands), and partner directly with our founders and domain experts to translate messy construction knowledge into robust backend systems.

What You'll Do

• Build and own backend services in Python - a FastAPI service backed by LangGraph pipelines that orchestrate multi-step agent reasoning with Postgres-backed checkpointing.
• Design and strengthen the agent pipeline - using vector search over a corpus of extracted detail views (Weaviate), embeddings, etc.
• Invest in evaluation - extend our eval framework so every change is measured, not guessed. Evals are a first-class deliverable here.
• Integrate LLMs pragmatically - work across multiple providers (Anthropic, OpenAI, Bedrock, Google, Groq, Ollama) via LangGraph, tuning prompts, structured outputs, and cost/latency tradeoffs.
• Shape architecture - write and review ADRs, weigh coupling and extension cost, and keep the system easy to extend without rewrites.

Our Stack

• Language: Python
• Services: FastAPI
• Agent orchestration: LangGraph, LangChain (core + multi-provider integrations)
• Data: PostgreSQL (psycopg), Supabase, Weaviate (vector DB)
• LLMs: Anthropic, OpenAI, AWS Bedrock, Google Vertex/GenAI, Groq, Ollama
• Infra: Docker / Compose, AWS (S3, ECR), GitHub Actions CI/CD
• Integration: a Revit plugin (in C#) that executes the placement commands our service generates

What We're Looking For

• 5+ years building and operating production backend systems, with strong Python depth.
• Experience designing API services and asynchronous/multi-step pipelines (FastAPI, or equivalent frameworks).
• Solid grasp of data modeling and databases relational (PostgreSQL) and ideally vector/search systems.
• A track record of owning systems end to end: design, ship, measure, iterate.
• Comfort working in an evaluation-driven culture - you reach for metrics and test harnesses, not vibes, to know if something improved.
• Strong engineering judgment: you weigh tradeoffs, write clearly about decisions, and keep systems simple and extensible.
• Clear written communication and the ability to collaborate closely with founders and domain experts.

Bonus Points

• Hands-on experience with LLM applications RAG/retrieval, embeddings, prompt engineering, structured outputs, agent frameworks (LangGraph/LangChain).
• Experience with retrieval/search quality work: ranking, recall/precision tuning, kNN, hybrid search.
• Cloud & infra fluency: AWS (S3, ECR), Docker, CI/CD.
• Background in AEC, CAD/BIM, computational geometry, or Revit or genuine excitement to learn a rich physical domain.
• Experience building or maintaining evaluation frameworks for ML/LLM systems.

Compensation

Above-market salary + meaningful ESOP.

Why Join

• Work on a genuinely hard, non-commodity problem - turning building-science expertise into a reliable AI system, not another chatbot wrapper.
• High ownership and direct impact; small team, fast iteration, real architectural influence.