New opportunity

Lead Architect: Product Engineer

Fractal Analytics Ltd. · Bengaluru, India

About this role

The Lead Architect: Product Engineer role for Cogentiq I2C—Fractal’s agentic AI product for invoice-to-cash—requires owning the engineering side of the application, services, data, and platform while partnering with an AI lead. You will design and build technical infrastructure, lead engineering execution, and write production code directly.

Responsibilities

Own technical design of the engineering side, including foundational platform decisions.
Drive delivery across the front end, service layer, data tier, and underlying infrastructure.
Coordinate and define interfaces between the engineering platform and the agent runtime jointly with the AI Lead.
Maintain technical quality through code review standards, rigorous testing, performance tuning, and practical trade-offs.
Ensure engineering health covering CI/CD, environments, release processes, technical debt management, and incident response.
Formulate the engineering plan by breaking the roadmap into sprint work, estimating realistically, and flagging slippage early.
Lead the engineering team by setting standards, coaching engineers, and keeping the team unblocked.
Stay hands-on in the code by taking on complex implementation slices to establish patterns for others.
Collaborate with product management on feasibility, sequencing, and trade-offs.

Required qualifications

Ten to fourteen years of software building experience, with at least three years leading a team or a full product build.
Proven track record of owning technical design for a product from scratch.
Deep expertise in at least one technical tier with working command across all tiers to lead, review, and debug.
Strong proficiency in Python with production experience using FastAPI or an equivalent service framework.
Practical experience with modern front-end tech like React or Next.js.
Data engineering depth including Spark, Databricks, Delta Lake, SQL, and sound judgement regarding lakehouse versus PostgreSQL storage.
Understanding of LLM and agentic systems to build supporting platforms and execute design discussions.
Product engineering track record building software for multiple clients rather than single engagements.
Working engineering discipline covering version control, CI/CD, automated testing, environment management, and structured incident handling.
Cloud experience on Azure or an equivalent platform.

Preferred qualifications

Order-to-cash or accounts receivable domain knowledge with familiarity with SAP FI-AR or equivalent.
Enterprise integration experience including ERP connectors, schema drift, and reconciliation.
Multi-tenant platform experience including access control and tenant separation.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonFastAPIReactNext.jsSparkDatabricksDelta LakeSQLPostgreSQLAzureCI/CDLLMAgentic SystemsSAP FI-AR

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Technical Design and Architecture of the Engineering Platform
As the engineering lead for Cogentiq I2C, you must own the end-to-end technical design covering the Next.js front end, FastAPI service layer, data tier on Azure Databricks with PostgreSQL, and underlying infrastructure, making foundational decisions that shape future scalability.
Collaboration with the AI Lead on Runtime Interfaces
Because the product splits leadership equally with an AI Lead owning agents and models, you must jointly establish clean, reliable interfaces between your engineering platform and the agentic runtime.
Hands-On Coding and Technical Leadership
The role requires staying active in the code by taking on the hardest implementation slices yourself, setting coding patterns, conducting code reviews, and mentoring the engineering team to maintain high delivery standards.
Data Engineering and Lakehouse Strategy
You are required to demonstrate data engineering depth across Spark, Databricks, Delta Lake, and SQL while exercising sound judgment on what data belongs in the lakehouse versus PostgreSQL.

Likely challenges to think through

Balancing Hands-On Coding with Engineering Management
Difficulty: Maintaining rigorous personal coding output while simultaneously leading a team, setting architectural standards, coaching engineers, and managing sprint planning. Preparation: Demonstrate a structured approach to delegating routine tasks, establishing clear team coding patterns through complex initial slices, and protecting time for architecture.
Managing Interfaces Between Deterministic Engineering and Probabilistic AI
Difficulty: Aligning traditional software tiers (FastAPI, Next.js, PostgreSQL) with a dynamic agentic AI runtime managed by a co-lead. Preparation: Prepare to discuss strategies for contract-driven API design, asynchronous communication patterns, and joint troubleshooting across boundaries where deterministic code meets probabilistic agent behavior.
Data Tier Architecture Decisions Between Lakehouse and Operational Database
Difficulty: Deciding structural boundaries between Azure Databricks Delta Lake and PostgreSQL without introducing latency or data integrity issues. Preparation: Review enterprise patterns for partitioning analytical workloads on Spark/Databricks from transactional relational data in PostgreSQL.

Practice your answers

Technical practice

Designing Interfaces Between Application Services and Agentic Runtimes

What to coverAs suggested preparation, discuss how you establish clean contract boundaries, asynchronous messaging, and state synchronization between a deterministic FastAPI service layer and an autonomous agent runtime. Address error handling, timeout management, and schema validation across component borders.

Data Partitioning Between Lakehouse and Relational Database

What to coverAs suggested preparation, explain your criteria for deciding whether high-volume analytical and financial data belongs in Azure Databricks Delta Lake versus transactional state in PostgreSQL. Highlight performance, concurrency, and schema evolution considerations.

Production FastAPI and Python Asynchronous Performance

What to coverAs suggested preparation, walk through how you optimize FastAPI application throughput under heavy concurrent load. Mention dependency injection patterns, async I/O handling, connection pooling, and profiling bottlenecks.

Spark and Databricks Pipeline Reliability and Delta Lake Features

What to coverAs suggested preparation, explain how you leverage Delta Lake ACID transactions, time travel, and schema enforcement to ensure data pipeline reliability in financial or enterprise data engineering workflows.

Modern Frontend Architecture Oversight with Next.js and React

What to coverAs suggested preparation, outline how an engineering lead without writing primary frontend code daily can maintain architectural standards, code quality, and performance best practices in a Next.js application.

Cloud Infrastructure and Multi-Tenant Platform Isolation on Azure

What to coverAs suggested preparation, describe strategies for implementing secure multi-tenant data isolation, access control, and tenant separation within cloud environments like Azure.

CI/CD Pipeline Design, Automated Testing, and Release Engineering

What to coverAs suggested preparation, explain how you structure automated testing suites (unit, integration, end-to-end), environment promotion, and zero-downtime release processes to safeguard engineering health.

Enterprise Integration, Schema Drift, and Reconciliation Patterns

What to coverAs suggested preparation, discuss how you handle schema drift and data reconciliation when integrating with external enterprise systems and ERP connectors.

Behavioural practice

Balancing Hands-on Coding with Team Leadership and Delivery

What to coverAs suggested preparation, prepare a situation/action/result example where you successfully balanced taking on the most complex coding tasks yourself while leading a team, unblocking engineers, and driving architectural standards.

Negotiating Technical Trade-offs and Roadmap Commitments with Product Management

What to coverAs suggested preparation, describe a time you worked with product managers to evaluate feature feasibility, sequence engineering work, and make deliberate trade-offs to meet external commitments without accruing unmanageable technical debt.

Resolving Architectural Disagreements Across Co-Leads or Cross-Functional Teams

What to coverAs suggested preparation, share an example of collaborating closely with a peer leader (such as an AI co-lead) to resolve differing technical perspectives, establish a unified design, and deliver a complex product component.

Managing Project Slippage and Proactively Mitigating Engineering Risks

What to coverAs suggested preparation, discuss an instance where an engineering plan or sprint was at risk of slipping, how you identified the bottleneck early, adjusted estimation, and communicated transparently with stakeholders.

Sharpen your knowledge

Topics to prepare

Agentic AI System Architecture and Runtime Interfaces
Conceptual topic: Understanding how deterministic application layers interact with probabilistic LLM and agentic runtimes. Exercise: Draft an architectural contract defining clear boundaries, payload formats, and error handling between a FastAPI service layer and an autonomous agent runtime. Relevance: Directly addresses the posting requirement to coordinate interfaces with the co-lead owning the AI side of Cogentiq I2C.
Lakehouse versus Operational Data Store Trade-offs
Conceptual topic: Evaluating data storage boundaries between Azure Databricks Delta Lake and PostgreSQL. Exercise: Map out an invoice-to-cash data flow determining which analytical workloads belong in Delta Lake versus transactional state in PostgreSQL. Relevance: Aligns with the mandatory qualification requiring sound judgment on lakehouse versus PostgreSQL storage.
Multi-Tenant Platform Architecture and Tenant Separation
Conceptual topic: Enterprise multi-tenancy, tenant isolation, and access control models. Exercise: Design a tenant isolation strategy covering data partitioning and role-based access control for a multi-client product. Relevance: Matches the preferred qualification for multi-tenant platform experience.
Order-to-Cash Domain and Enterprise Integration Patterns
Conceptual topic: Business processes for collections, cash application, deductions, and ERP integrations. Exercise: Review enterprise integration patterns for handling schema drift and ERP connectors. Relevance: Relates to the preferred qualification in order-to-cash domain knowledge and SAP FI-AR familiarity.

Skills to revise

Python and FastAPI Production Service Design
Hands-on skill: Building robust, scalable microservices and REST APIs in Python using FastAPI. Exercise: Implement a production-grade FastAPI endpoint featuring asynchronous request handling, input validation with Pydantic, structured logging, and automated testing. Relevance: Directly fulfills the requirement for strong Python and production experience with FastAPI.
Azure Databricks and Spark Data Engineering
Hands-on skill: Processing large-scale data pipelines using Spark, Delta Lake, and Databricks. Exercise: Write a PySpark job using Delta Lake features (such as time travel and ACID transactions) to process and reconcile batch financial data. Relevance: Aligns with the core data engineering depth requirement.
Modern Frontend Oversight with Next.js and React
Hands-on skill: Reviewing and guiding modern frontend application architectures. Exercise: Audit a Next.js/React component structure for performance bottlenecks, state management efficiency, and API integration patterns. Relevance: Supports the requirement for practical experience with Next.js and React to hold engineering standards.
CI/CD Pipeline and Engineering Health Automation
Hands-on skill: Configuring automated testing, version control workflows, environment management, and release processes. Exercise: Build a GitHub Actions or Azure DevOps pipeline that automates linting, unit testing, integration testing, and containerized deployment. Relevance: Matches the requirement for working engineering discipline and CI/CD maintenance.

Your preparation checklist

Prepare a Multi-Tier Architecture Walkthrough
Assemble a portfolio artifact or system design diagram illustrating a product you designed that spans a modern frontend (React/Next.js), service layer (FastAPI/Python), and data tier (lakehouse and relational database), demonstrating end-to-end technical ownership.
Draft an Agent Runtime Integration Contract
Create a sample interface specification document outlining how deterministic application microservices coordinate API payloads, error handling, and state synchronization with an autonomous LLM/agent runtime.
Compile Data Engineering Decision Frameworks
Prepare notes and architectural rationale detailing when to store analytical and financial datasets in Azure Databricks Delta Lake versus transactional state in PostgreSQL, highlighting Spark performance and lakehouse trade-offs.
Structure Code Review and Engineering Health Standards
Draft a concise checklist or guideline document covering code review standards, CI/CD pipeline automation, test coverage thresholds, and structured incident response procedures to demonstrate team leadership capabilities.
Formulate Thoughtful Employer Questions on Product Co-Leadership
Prepare specific questions for the interviewers regarding how engineering co-leads divide responsibilities with the AI lead on Cogentiq I2C, manage joint release cycles, and collaborate with product management on roadmap sequencing.

Availability has not yet been verified. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…