Manish_Gupta_Tpgl

AI is rapidly transforming the Reverse Supply Chain (RSC)—spanning returns, repair, refurbishment, resale, recycling, and compliance. Over the next 5–10 years, entire career paths will evolve as analytical, operational, and product roles increasingly integrate AI-driven decisioning and automation. Below is a streamlined view of how this progression may unfold and the capabilities that will define success at each stage. 1) Today–2 Years: Returns & Circularity Analyst Scope: Returns forecasting, grading variance analysis, RMA policy interpretation, vendor chargebacks, disposition mix optimization (restock/refurbish/recycle), and warranty cost reporting. AI Impact: ML-driven demand and returns forecasting (seasonality, promotions, channel behaviors). Computer vision for automated condition grading and fraud detection using inbound images/video. LLM copilots to summarize disposition decisions, extract policy clauses, and draft vendor disputes. Optimization models for selecting the most profitable and cost-efficient disposition pathway. Key Skills: SQL, Python, BI tools. Introductory ML (time-series, classification). Knowledge of returns/warranty policies and EPR/e-waste regulations. 2) 2–4 Years: Analytics Lead / Data Product Analyst Scope: Leads model ownership, dashboard development, grading standardization, and collaboration across operations and finance. AI Impact: Digital twins to simulate reverse flows, capacity, SLAs, and cost-to-serve. Policy mining via LLMs to highlight non-compliance and margin leakage. Dynamic, network-aware routing to improve recovery value and turnaround time. Anomaly detection for fraud patterns, warranty abuse, and hidden defects. Key Skills: Model deployment and monitoring (feature stores, drift detection). Decision science and optimization methods (e.g., linear programming). Data governance, PII handling, and explainability documentation. Agile ways of working: user stories, backlogs, release management. 3) 4–6 Years: Product Owner (AI & Automation) Scope: Owns the end-to-end Returns Decisioning Platform—from intake to disposition to resale activation. AI Impact: Multi-objective optimization balancing profit, carbon footprint, and service levels. Autonomous workflows: automated RMAs, label selection, and instant refunds for low-risk segments. Intelligent resale orchestration: automatic listing creation, pricing, and channel selection. Automated sustainability reporting: Scope 3 data and end-of-life insights. Key Skills: P&L and sustainability metrics. AI governance, risk controls, and human-in-the-loop design. Vendor and partner ecosystem management (3PLs, refurbishers, marketplaces). Change management and narrative building for adoption and ROI. 4) 6–8 Years: Circularity Operations Manager (Network) Scope: Leads a multi-node refurb/repair/recycling network with responsibility for budgets, scorecards, and regional alignment. AI Impact: Network digital twins enabling near–real-time policy adjustments (surge returns, part shortages). Predictive maintenance for repair and testing infrastructure. Carbon-aware routing and incentives to support ESG goals. Key Skills: Network design, vendor contracts, and operational incentives. Scenario planning for recalls, regulatory shifts, and supply constraints. Strong financial management: recovery margins, cash flow, write-off prevention, and aging control. 5) 8–10 Years: Business Head - Reverse Logistics Scope: Manage complete P&L AI Impact: Enables closed‑loop feedback to design and sourcing teams through product defect intelligence. Supports dynamic business models (trade‑in, buy‑back, recommerce) through advanced risk and pricing analytics. Automates compliance with evolving EPR, right‑to‑repair mandates, and cross‑border waste regulations. Key Skills: Enterprise‑level collaboration and strong partnerships with OEMs, regulators, and marketplaces. Executive and external stakeholder management, with strong communication and data storytelling capabilities. Deep understanding of governance, ethics, and responsible AI practices. What Will Drive Progression & Advancement 1) Decision Quality & Measurable Impact Early roles: accuracy (grading, forecasting), cycle time, recovery %, SLA performance. Senior roles: network margin uplift, cash flow impact, carbon reduction, policy leakage elimination, and recommerce revenue growth. 2) AI Product Thinking Framing operational problems as scalable decisioning services. Defining datasets, labels, human-in-loop checkpoints, and continuous feedback loops. 3) Optimization & Systems Skills Multi-objective optimization (profit, carbon, customer experience). Simulations and digital twins for scenario planning. 4) Controls, Compliance & Explainability EPR, WEEE, battery rules, WPC/ETA (India), and right-to-repair. Ability to document model lineage, bias checks, and audit-ready evidence. 5) Partner Ecosystems & Platformization Building integrated APIs, scorecards, and SLA frameworks for 3PLs and refurb partners. Performance-linked contracts leveraging AI metrics. 6) Leadership & Change Enablement Cross-functional collaboration with CX, Finance, Legal, and Sustainability. Upskilling teams and building operating rhythms around AI insights. KPIs That Will Gain Importance Net Recovery Margin (post all costs). Time-to-Value (receipt → resale/repair completion). Grading accuracy and consistency (CV-assisted). Disposition uplift vs. rule-based benchmarks. Sustainability metrics: carbon per returned unit, landfill diversion, % parts harvested. Policy leakage closure across warranty and vendor recovery.

As AI systems increasingly handle data-heavy, repetitive, and analytical tasks in reverse logistics — such as identifying return patterns, forecasting refurbishment demand, or automating disposition recommendations — the role of humans shifts from performers to judges, framers, and orchestrators. Traditional requirement: • Strong Excel/ERP skills • Ability to manually process and clean data • Knowledge of return codes, material movement, SLAs • Ability to create static reports and forecasts • Basic problem-solving around exceptions These reflect the rising need for human judgment, oversight, and business framing. 1. Systems Thinking & Process Understanding 2. Critical Judgment & Exception Handling 3. Data Interpretation & Storytelling 4. Collaboration & Change Management Qualities That Become Less Critical 1. Raw Data Processing Skills e.g. manual Excel crunching, pivot-heavy reporting etc 2. Memorizing Return Codes or Policies AI systems can: auto-suggest disposition, surface relevant warranty rules, detect mismatches in claim reasons, Knowledge still matters — but recall speed becomes less critical. 3. Repetitive Operational Execution Roles that used to rely on: high-volume RMA processing, manual case triage In Reverse Supply Chain Operations: As AI takes over analysis, pattern recognition, and rule-based decision-making, recruitment must prioritize: • Judgment • Systems thinking • Interpretation & business framing • Ethical oversight • Human-AI collaboration And rely less on: • manual data work • memorization-heavy tasks • repetitive operational processing

In Sales Operations, the initial model that we deployed was a complex one as it was a combination of three processes. Step 1. Autogenerating Descriptions from "one to many" and “many to many” relationship structure Step 2. Auto QA Step 3. Auto upload the correct answer if QA is good enough. Original (pre-AI) workflow The earlier (pre‑AI) workflow was entirely manual, with each process operating independently. It required extensive human effort, was prone to errors, and Lacked scalability due to high dependency on manual intervention. The performance expectation was that all metrics should consistently meet or exceed the defined SLAs. AI workflow After substantial learning and refinement, we were able to simplify the complex “one‑to‑many” and “many‑to‑many” structures. We formed a dedicated team to ensure that any issues or fallouts were identified and resolved at an early stage. QA: All outputs were automatically mapped to the desired target state. Initial fallouts were thoroughly revalidated to ensure accurate root‑cause analysis. Subsequently, we developed an automated upload mechanism that leveraged the outputs from the above processes, effectively making the workflow “self‑correcting.” The inherent AI risks and hidden errors primarily came from the “garbage in, garbage out” principle. It took several quarters of learning and refinement to improve data quality and model outputs. During the period, human effort increased significantly, as teams had to review the AI‑generated outputs in addition to managing their regular responsibilities. New skills that came to forefront were: Developing deep Subject Matter Expertise (SME). Collaborating with AI systems to guide learning and improvement. Demonstrating greater resilience during iterative refinements. Continuously monitoring and validating final outputs for accuracy. Traditional skills becoming less relevant: Performing repetitive, manual tasks. Following labor intensive work methods as faster ways are easily available. Producing static, one dimensional outputs. Now everyone is looking for two-way communication. Managing slow and time consuming processes. Performance metrics would still be aligned to the business but we can expect more accuracy, faster cycle times, less human errors and more scalability. Training intervention that works: Change Management trainings on how to accept the changes Skill updation to make create SME force Tool trainings/ On the Job training Share real historical user cases and how businesses & employees were able to benefit from the same. Allow humans to challenge the output and get their help for RCAs. Issue identification training - Compare, discuss differences, and explain reasoning Prompts & Interpretation Training

In Operations domain, we’ve deployed few AI applications. Although we hope that AI results are always better than manual work but there have been times when AI output has been not up to the mark. In traditional Operation’s work, lot of manual or automated work has been done on daily basis. Tasks are normally repetitive but time and again new scenarios come up which requires human inputs. Metrics normally depends on what kind of process are we running in the organization. In traditional metrics environment, we go for balanced scorecard i.e. Production and People aspect example in Reverse Supply Chain, - Units Received - Units WIP - Units in Ready State - Units Sold - Margin - Man hours utilized - QA And few more With AI, although the business metrics will not change much but we will see improvements across the board. Example: - Received, WIP & Ready State cycle time should be reduced - Man-hours utilized should show significant changeover. - Error count in QA should be reduced by significant level The behaviors that we should encourage - Monitor the AI output and ensure its accuracy - Make callouts early so that damage can be controlled at the earliest. - Look for opportunity areas to implement AI. - Check for Cycle time benefits - Cost benefit of implementing AI

We need people to trust their expertise and judgement while working with AI tools. AI will provide lot of information in a digestible format which makes it look good to believe in the content. We should look for Historical references - if data / information provided by AI has been valuable or not. Human in the loop - Cross check the information with Subject Matter Expertise or other relevant tools. If data provided is not of good quality or there are loopholes, then review the output before putting it to use. In scenarios where historical data is not available then AI models will not work accurately. If we believe data provided to AI was good and output is in line with what we were expecting, then we can utilize the output.