Paul Iusztin

🤖 Senior Machine Learning Engineer • MLOps • Contractor ~ Helping startups engineer production-ready ML/AI systems.

I'm the author of the bestseller LLM Engineer's Handbook, lead instructor of the Agentic AI Engineering course, founding AI Engineer of a San Francisco start-up, and obsessed with making knowledge accessible through AI.

With over 10 years of experience in AI and software and 20 apps shipped, I teach AI Engineering as I wanted to at the beginning of my career. End-to-end. From idea to production. From data collection to deploying, monitoring and evaluation. With a focus on AI principles, software patterns and infrastructure systems that will thrive in a future dominated by AI coding tools.

My ultimate goal is to help other engineers escape PoC purgatory and 10x their AI Engineering skills.

.

If you want to become an AI Engineer pro, join Decoding AI for exclusive content on designing, building, and shipping AI software that works: 

🗞️ https://www.decodingai.com

See more:

🌐 Homepage: https://www.pauliusztin.ai
📚 LLM Engineer's Handbook: https://www.amazon.com/LLM-Engineers-Handbook-engineering-production/dp/1836200072/
🤖 Agentic AI Engineering Course: https://academy.towardsai.net/courses/agent-engineering?ref=b3ab31

.

✉️ For collabs, contact me at: p.b.iusztin@gmail.com

Senior Machine Learning Engineer

🟡 Scalable ML batch infrastructure for continuous training & inference

The client, a generative AI leader, struggled with manual training and calling their ML models. As a tedious and unscalable process, they needed to automate their ML processes completely and integrate them with their current web platform.

➞ 𝐃𝐞𝐬𝐢𝐠𝐧𝐞𝐝 and 𝐢𝐦𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐞𝐝 a modular 𝐏𝐲𝐭𝐡𝐨𝐧 𝐩𝐚𝐜𝐤𝐚𝐠𝐞 that composes the ML model required steps into 3 main batch pipelines that can easily be automated: feature, training, and inference, 𝐞𝐥𝐢𝐦𝐢𝐧𝐚𝐭𝐢𝐧𝐠 a 𝐭𝐞𝐝𝐢𝐨𝐮𝐬 𝟐-𝐡𝐨𝐮𝐫 𝐦𝐚𝐧𝐮𝐚𝐥 𝐩𝐫𝐨𝐜𝐞𝐬𝐬/𝐯𝐢𝐝𝐞𝐨.
➞ 𝐃𝐞𝐩𝐥𝐨𝐲𝐞𝐝 a scalable & cost-effective 𝐛𝐚𝐭𝐜𝐡 𝐚𝐫𝐜𝐡𝐢𝐭𝐞𝐜𝐭𝐮𝐫𝐞 and 𝐂𝐈/𝐂𝐃 𝐩𝐢𝐩𝐞𝐥𝐢𝐧𝐞 on top of AWS ECS & SQS that scales in and out based on the number of jobs, 𝐬𝐚𝐯𝐢𝐧𝐠 𝐭𝐡𝐨𝐮𝐬𝐚𝐧𝐝𝐬 𝐨𝐟 𝐝𝐨𝐥𝐥𝐚𝐫𝐬/𝐦𝐨𝐧𝐭𝐡.
➞ 𝐒𝐚𝐯𝐞𝐝 𝐦𝐨𝐧𝐭𝐡𝐬 𝐨𝐟 𝐟𝐮𝐭𝐮𝐫𝐞 𝐝𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭 by using good software practices to make the Python module & batch cloud infrastructure flexible and composable.
➞ 𝐁𝐮𝐢𝐥𝐭 a 𝐯𝐨𝐢𝐜𝐞 𝐬𝐲𝐧𝐭𝐡𝐞𝐬𝐢𝐬 𝐏𝐨𝐂 in ~1 week required to 𝐜𝐥𝐨𝐬𝐞 the 𝐝𝐞𝐚𝐥 with one of the 𝐜𝐮𝐫𝐫𝐞𝐧𝐭 𝐜𝐥𝐢𝐞𝐧𝐭𝐬.

⚫️ Big data ingestion & archiving pipeline

The client wanted to expand its platform features by securely and safely ingesting, preprocessing, and storing TB of data into their platform.

➞ 𝐂𝐨𝐧𝐭𝐫𝐢𝐛𝐮𝐭𝐞𝐝 to the 𝐝𝐞𝐬𝐢𝐠𝐧 and 𝐢𝐦𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 of a 𝐝𝐚𝐭𝐚 𝐢𝐧𝐠𝐞𝐬𝐭𝐢𝐨𝐧 𝐩𝐢𝐩𝐞𝐥𝐢𝐧𝐞 that supports 𝐢𝐧𝐠𝐞𝐬𝐭𝐢𝐧𝐠 >𝟓 𝐓𝐁 of data in 𝐨𝐧𝐞 𝐠𝐨 in a reliable, fast, and secure way.
➞ 𝐈𝐦𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐞𝐝 a scalable 𝐝𝐚𝐭𝐚 𝐩𝐫𝐞𝐩𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠 𝐩𝐢𝐩𝐞𝐥𝐢𝐧𝐞 that supports multiple media types (images, videos, audio, 3D objects) and creates small preview files that are cheap to transfer, store and visualize. The 𝐨𝐫𝐢𝐠𝐢𝐧𝐚𝐥 𝐟𝐢𝐥𝐞𝐬 are 𝐚𝐫𝐜𝐡𝐢𝐯𝐞𝐝 on an affordable storage solution 𝐫𝐞𝐝𝐮𝐜𝐢𝐧𝐠 𝐜𝐨𝐬𝐭𝐬 𝐛𝐲 𝐮𝐩 𝐭𝐨 𝟖𝟏%.

I built a system that turns a single idea into a publish-ready technical article.

Here's a mini masterclass on how (so you can copy):

The system consists of 2 parts:
 
• Nova (research agent): exploratory, steerable
• Brown (writing workflow): deterministic, auditable

𝟭) 𝗡𝗼𝘃𝗮: 𝗦𝘁𝗲𝗲𝗿𝗮𝗯𝗹𝗲 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗔𝗴𝗲𝗻𝘁 (𝗲𝘅𝗽𝗹𝗼𝗿𝗮𝘁𝗶𝗼𝗻)

Goal: ingest context from messy sources, fill gaps, and produce a clean research artifact.

Inputs:
 
• Structured article brief (topic, audience, structure, constraints, candidate sources)

How it runs:
 
• MCP-first design so tools are portable across harness clients
• A prompt exposed as procedural memory to plan and glue all stages
• Tooling focused on ingesting context, not “thinking in tokens”

Core stages:
 
• Pull context from URLs, local files, GitHub repos, YouTube transcripts
• Run multi-round research loops (Perplexity for cited web search)
• Filter for trust and relevance
• Deep scrape the top sources
• Consolidate into a single handoff artifact: research.md

Output:
 
• A curated research document with organized notes and citations, ready for drafting

𝟮) 𝗕𝗿𝗼𝘄𝗻: 𝗥𝗲𝗹𝗶𝗮𝗯𝗹𝗲 𝗪𝗿𝗶𝘁𝗶𝗻𝗴 𝗪𝗼𝗿𝗸𝗳𝗹𝗼𝘄 (𝗲𝘅𝗲𝗰𝘂𝘁𝗶𝗼𝗻)

Goal: turn research into a consistent article, with quality gates and checkpoints.

Why a workflow, not an agent?

Because writing quality demands:
 
• Repeatability
• Checkpoints
• Clear stage boundaries
• Traceability when something is off

How it runs:
 
• LangGraph for stateful orchestration, resumability, and auditability
• Leveraging tools as deterministic workers instead of exposing them to agents

Core stages:
 
• Load context: guideline + research.md + writing profiles
• Generate required assets in parallel (diagrams, images, tables when needed)
• Draft the article
• Run review and rewrite loops (reflection scores, style judge, revision triggers)
• Support human edits without breaking the system’s rules

Outputs:
 
• Final article
• Generated assets + metadata
• Reflection scores and traces (so you debug quality instead of guessing)

Here’s the gist…

If you promise “production-grade writing”, you need exploration for research and durability for execution.

This design choice is the difference between a cool demo and a system you can ship.

Want to learn how to build it out yourself?

We break this down step-by-step in the Agentic AI Engineering course (built with Towards AI, Inc.):

Enroll here: https://lnkd.in/dtxZuZpb

P.S. We’ve opened up another 100 seats for $499. The last batch sold out in under 14 days → get in early to lock in our best offer.

Senior AI Engineer • Founder @ Decoding AI • Author @ LLM Engineer’s Handbook ~ I ship AI products and teach you about the process.

aecc - digital innovation hub

We’re reimagining the future of international student education through transformational change by merging data, creativity & and technology to create brilliant customer experiences. 

The average student's journey to take a life-defining career choice to study abroad varies from 120 days to 480 days, right from creating a profile to shortlisting and finalizing a university is an exhaustive process. 

I primarily work on AECC Referrals products, which include Student Referral (B2C) and Merchant Referral (B2B) areas of business, a product that is a direct and authentic way to spread the word about our services, motivating and engaging existing students to share our customer delight with their network.

Students share their personal experiences with friends and family. They are our brand ambassadors to others, sharing their same needs and pain points, and therefore in the market for our offerings.

AECC Referrals is a tiered reward/incentive program, that gives students and our alumni a token of thanks for referring others, fostering trust for other products and services we offer. Student Referral helped increase customer engagement, advocacy & more excellent lifetime value. The product also translated into an effective ROI strategy with a strong word-of-mouth marketing effect. 

Business Outcomes - 

a. Improved Student Referrals with an 83% increase in leads generated. 

b. Improved conversion rate by 33% resulting in a Q2 revenue of 4 million rupees.

Product Manager

Incrementum Digital

Catalog Catalyst 

Developed a Product Experience Management SaaS software (0 to 1 build) that centralizes, synchronizes, and integrates product information with every component of a commerce technology stack such as e-commerce software, 3rd party software & plugins allowing each layer to exchange data smoothly.

Provides e-commerce merchants & resellers with actionable product purchasing insights & ROI.

Fragmentation of e-commerce operations, marketing, sales, website(s), and product information - all of those functions built into an omnichannel operations platform. 

Catalog Catalyst is customized for the organization's optimization of sales velocity, and to achieve high revenue multiples by enabling them to:

· Efficiently deal with complex, bulk data

· Manage digital assets

· Create multiple catalogues and categories

· Build a large number of SKUs

· Improve product presentation

· Deliver personalized shopping experience.

Spicywagon

Developed a 3-sided Marketplace Platform, Growth Engine, Information Management and Operations design to scale to 410 railway stations across India.

Devised effective Go-to-Market Strategies through the 'Group Bulk Order' feature to be able to increase 'Average Order Value' by 77%.

Developed low-fi Customer Ordering App in Figma with features - a. Find & Locate Train/PNR b. Restaurant Listings c. Restaurant Pages d. Cart e. Destiny Place Delicacies (Ethnic food discovery page of itinerary stations) f. Order Tracking 

Secured Indian Railway Catering and Tourism Corporation (IRCTC) 3+1 years official 'E-Catering for Aggregators' contract to facilitate food in trains across 900+ railway stations pan India.

Owned 'Spicywagon CRM Dashboard' and 'Vendor Partner App', to increase 'Closure of Orders' by 40% and Vendor Partnerships by 70%.

Developed 'Smart' and 'Batch' Orders Notifications, and ‘Proof-Of-Delivery’ features as part of the Product Roadmap to handle demand surge.

Developed Graph-based Trains Information System, which enabled live train tracking thereby increasing the 'Order Delivered' success rate from 82% to 93%.

Launched - 'Commission Structure Modelling' and 'Dynamic Menu Pricing' features on the internal Spicywagon dashboard.

Growth Generalist.

Principal Contributor - Business modeling for the 3-sided marketplace, Market & Take Rate Analysis.

Product Growth

Spicywagon is an on-demand food aggregator and delivery startup in Indian Railway trains, we list restaurants nearby stations throughout the journey itinerary of passengers and deliver the food right to the passengers' seats. 

Secured 100K USD of pre-seed funds towards business development & operations with a GTM strategy, targeting 50 high-performing railway stations to kick start as aggregators on the IRCTC 'E-Catering Platform' (https://www.ecatering.irctc.co.in/).

Passengers can pre-book food through our service online and order it on the train throughout their journey. 

The observations of the below article inspired the startup idea -  

"CAG India, slams Indian Railways, says 'food unfit for consumption' being sold at railway stations"

[https://cag.gov.in/uploads/download_audit_report/2017/Report_No.13_of_2017_-_Performance_audit_on_Catering_Services_Reports_in_Indian_Railways_Union_Government_Railways_0.pdf]

Co-Founder

Birlasoft

Techno functional consultant working in Oracle PaaS EBS HCM domain, HR, Payroll, Time & Labor, Projects, Advanced Benefits, Self - Service HR and Talent Management modules.

Increased the Annual Contract Value of Birlasoft-AMRI clientele by USD 100K by introducing Oracle OTL to effectively capture their contract workers effort.

Attended industry seminars and training sessions to promote products and services to clients.

Customer Success 

Tech Mahindra

Functionally designed, implemented, support complex SaaS solutions in Oracle EBS HCM closing over 1000 incidents/tickets in the support over 3.5 years with less than 0.2% SLA breach.

Initiated pre-sales activities, consulting services and products – Developed projects’ scopes (statement of work, scope of work) and prepared proposals to win USD 10 Million Balfour Beatty PFG (Platform for Growth) Cloud EBS Project.

Implementing research activities to analyse client feedback and monitoring new, existing project plans to implement legacy to cloud system transition for Balfour Beatty Client.

Oracle HCM Consultant

Product Growth | Ex-Founder | Football Enthusiast

Me in 10 secs:

- I'm a product guy, a startup founder, and a football analyst.
- Actively working in the AI eco-system, NLP-to-SQL problems, Open Source AI models and AI Agents. Learn more about it here - https://medium.com/@aniruddh.gilakamsetty/text-rag-to-sql-solution-for-journeyiq-ace079772737
- Ex-Product Manager at AECC. Previously built a 0-1 B2B SaaS product, Catalog Catalyst (Product Experience Management Tool) at Incrementum Digital.
- Previous to that, built a pre-seed funded ($100K), on-demand food delivery startup for the Indian Railway Trains eco-system, Spicywagon that grew from 0-10X, an official IRCTC authorized food aggregator. Learn more about it here - https://medium.com/@aniruddh.gilakamsetty/spicywagon-food-delivery-in-indian-railway-trains-teardown-part-1-44c43a018b64

Me in the timeline: A 13+ years of professional experience is as follows, 

2024+
- Built a Text-to-SQL Gen-AI engine that converts natural language questions (plain simple english) into corresponding SQL queries that can be executed on relational databases.
- Joined the masters program at Lund University, Sweden.

2023+ 
- Product manager at AECC - Digital Innovation Hub, an international overseas education consultancy primed to solve various student-centric problems to study in their dream university. Currently solving for Referrals, a strong e-WORM to amplify a student's aha moment to peers, in the long and life-deciding journey to study aboard in their destination country.

2022 - 2023
- With Incrementum Digital, a B2B SaaS digital transformative service for clients to enhance their digital presence and performance.

From 2016 - 2021, 
- Pursued my interest in building and co-founding, a pre-seed funded USD 100K, B2C, on-demand food delivery startup in the Indian Railway Trains eco-system, Spicywagon. 

- Primary skills in Product Management, Product Strategy, Stakeholder Management, UI/UX Prototyping, Market Research, and Low-Code/No-Code Development.

2011 - 2016 
- A combined 5.5 years of IT experience in Oracle HCM as a Customer Success SaaS Consultant in India’s leading IT consultancies Tech Mahindra Pvt Ltd. as a fresher – college campus placement followed by Birlasoft India Pvt Ltd.

taplio

The Battle

Paul Iusztin

Aniruddh Durga Naga Gilakamsetty

Followers

Posts

Engagements

Likes

Performance ScorePaul Iusztin's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Paul Iusztin

Aniruddh Durga Naga Gilakamsetty

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Paul Iusztin's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScorePaul Iusztin's LinkedIn Performance