AI Data Architect
Persistent Systems is seeking an experienced AI Data Architect to lead the design and delivery of scalable AI and LLM-driven enterprise systems. The role focuses on architecting production-grade applications, including chatbots and agentic workflows, while ensuring cost efficiency, performance, and security. Candidates are expected to have deep expertise in building AI platforms, integrating data ecosystems, and implementing robust governance. The position is based in Pune and offers a hybrid work environment.
50k new jobs listed every day. Install TAL to find more jobs like this.
Experience
12+ years
Function
Engineering
Work mode
Hybrid, India
Company
Tier 2
What you will work on
Persistent Systems is seeking an experienced AI Data Architect to lead the design and delivery of scalable AI and LLM-driven enterprise systems. The role focuses on architecting production-grade applications, including chatbots and agentic workflows, while ensuring cost efficiency, performance, and security. Candidates are expected to have deep expertise in building AI platforms, integrating data ecosystems, and implementing robust governance. The position is based in Pune and offers a hybrid work environment.
TAL's take
Senior architect role at a large, established IT services provider with clearly defined expectations and technical requirements.
The JD provides a very clear, professional scope of responsibilities and technical expectations for an AI architect.
Salaries at Persistent Systems
16.9 LPA average
Based on 906 Grapevine salary entries for Persistent Systems.
Engineering
0 - 2 years | 3.1
5 LPA average
Range: 4 - 8 LPA
Engineering
2 - 4 years | 3.1
7 LPA average
Range: 4 - 24 LPA
Engineering
4 - 6 years | 3.3
13 LPA average
Range: 7 - 25 LPA
Engineering
6 - 8 years | L5
16 LPA average
Range: 8 - 27 LPA
Must haves
- 12+ years of experience
- Designing and delivering production AI/LLM/chatbot systems
- Optimizing LLM cost, latency, and throughput
- Integrating structured data platforms and data warehouses
- Designing testing, evaluation, and validation strategies for LLM systems
- Implementing security controls, RBAC, and data access boundaries
Tools and skills
About the company
Established IT services firm with significant scale and global footprint, but lacks the specialized product-led status of tier 1 tech.
Posts mentioning Persistent Systems
I have two offers Of same CTC Capgemini and PERSISTENT SYSTEMS which one is good to join?
How much persistent give for 11 years exp salary and grade for a full stack developer and having deveops and cloudops knowledge
Title: Data Scientist (~2 YOE) – Built planning tools, pipelines, and AI system. Need honest feedback on profile.
Hi all, Looking for practical feedback on my profile before I start applying. I’ll keep this structured so it’s easier to evaluate. --- 1) Planning Tools / Web Applications Problem: Forecasting workflows were fragmented and heavily Excel-driven: - Multiple data sources (orders, shipments, different forecast versions) - Manual merging, lookups, and adjustments - No way to simulate scenarios or compare forecasts cleanly - Different planners using different methods → inconsistency What I built: - Two internal applications for planning workflows: - A planning tool integrating 8+ data sources - A forecasting simulator supporting multi-level editing (high → granular) Key capabilities: - Real-time scenario simulation - Side-by-side comparison of multiple forecast types - Hierarchical adjustments across levels - SQL write-back for persistence Scale: - Processes ~150K+ records per cycle - Used in monthly planning cycles by multiple teams Impact: - Removed fragmented Excel workflows - Enabled consistent decision-making across users - Reduced manual effort and improved visibility into forecast behavior --- 2) Automation & Data Pipelines Problem: Core workflows were manual and repetitive: - Multi-file Excel processing - Data cleaning + merging across systems - Version tracking errors - High effort per cycle (1–4 hours depending on workflow) What I built: - Multiple pipelines automating end-to-end workflows Examples: - Large-scale consolidation pipeline: - Input: ~1M+ rows across 20+ files - Output: clean, unified dataset (~75% reduction) - 2nd pipeline: - Replaced a 23-step manual process - Standardized inconsistent formats across datasets - 3rd automation processing: - Automated unpivoting, enrichment, and version tracking Impact: - Reduced processing time from hours → minutes per cycle - Eliminated manual errors (copy-paste, lookup mistakes) - Standardized workflows across users --- 3) Power BI / Monitoring Problem: Recent data (orders/shipments) showed inconsistencies, but: - No visibility into changes over time - Hard to identify where data drift was happening What I built: - Power BI dashboards with: - Hierarchical filters - Drill-down views - Month-over-month comparison Scale: - ~30K+ records analyzed Impact: - Enabled early detection of data inconsistencies - Helped planners validate inputs before forecasting - Improved trust in upstream data --- 4) Side Project (AI System) What I built: - AI-powered job assistant system Features: - Scrapes job postings - Scores relevance using LLMs - Generates tailored resume points and outreach messages - Tracks applications Tech: - FastAPI backend - LLM routing (cloud + local fallback) - SQLite storage Goal: - Build a system-driven workflow (not just model usage) --- My concern Most of my work sits at the intersection of: - forecasting - data systems - workflow automation I’m trying to move into: 👉 Applied Data Scientist / Product-oriented roles --- Questions 1. Does this profile look too niche (forecasting-heavy)? 2. Does “building systems around data” help or hurt for DS roles? 3. What’s the biggest gap you see (if any)? Would really appreciate honest feedback. Thanks.