Revanth Banala profile photo 1 on Tal

Revanth Banala

Founder, Lumere AI | Building neuroscience-informed adaptive learning | CMU

Bengaluru

4+ years of work experience

Lumere AI
Nov 20258 mosPresent

Bengaluru

AI for education — we help institutions understand how their students actually learn. Our engine detects specific misconceptions, tracks them across sessions, and gives teachers real-time visibility into who needs help and on what. Not "Rahul scored 34%." More like "Rahul has been confusing the same concept for 3 weeks and nobody's caught it." Started with JEE/NEET. Building toward K-12. Live product. Serving institutions. That's what we're building. For institutions. For students. At scale.
Coldbean.ai
Aug 20241 yr 1 moAug 2025

San Francisco Bay Area

Joined as an AI Strategy Advisor and later took on the full-time CTO role (while at UnifyCx), architecting Coldbean’s AI stack end-to-end. Built agentic systems for sales intelligence, enrichment, outreach, and onboarding. Scaled the tech team from zero, drove IP-grade innovation, and shipped multiple production-ready AI workflows. The journey eventually closed, but it cemented my conviction for solo-founder-led, execution-driven AI products — a philosophy now core to my next venture.
UnifyCX
Mar 20241 yr 6 mosAug 2025

Santo Domingo, Dominican Republic

Led critical initiatives across platform architecture, LLM-based agent design, and enterprise demos—blending deep technical execution with product strategy and research-backed innovation. My role spanned AI research, infra, internal tooling, and customer-facing solutions. Key Contributions: • Inventor of a Patent-Worthy Intent Platform: Single-handedly built a scalable alternative to systems like Amazon Lex—initially a scrappy side project with a group of interns. Now used across all deployments, it cut onboarding time from ~7 days to 20 minutes and is under patent consideration. • LLM Agent Architecture & Production Systems: Refactored core flows into modular, agentic systems capable of dynamic routing, follow-ups, and summarization. Drove platformization efforts to enable reuse across bots, backend services, and client-facing AI agents. • Enterprise-Grade PoC that lead to a $30M+ engagement: Designed and led delivery of a production-ready demo that directly contributed to a $30M+ enterprise engagement. The solution still anchors high-stakes executive pitches and live sales use cases. • Feature Lead for Study Buddy: Owned end-to-end ideation, design, and execution of a contact-center agent support tool with high engagement. Shaped AI capabilities across multiple agent-assist modules through hands-on delivery and technical strategy. • Product Co-Creation & Insight Mining: Worked closely with Product to define roadmap for features like Voice of Customer. Delivered insights from AI-driven mining that surfaced user pain points product teams hadn’t uncovered—earning praise from stakeholders. Beyond these, I consistently pushed platform thinking across teams—building internal tools like model registries, reusable config runners, agentic flows —accelerating AI deployment velocity org-wide.
UnifyCX
Mar 20231 yr 1 moMar 2024

Boston, Massachusetts, United States

Joined as the first AI hire and helped shape the foundation of Lightbird’s GenAI platform—driving hands-on development, research, and early team growth while wearing multiple hats across engineering, product, and client success. Key Contributions: • Built the First Wave of AI Modules: Designed, prototyped, and productionized critical systems—ticket categorization, intent recognition, QnA systems, product detection, and query enrichment—that still anchor the AI assistant today. • Hands-On Technical Leadership: Drove core AI infra from scratch—defining DTO standards, test frameworks, deployment templates, and prompt stacks—used as the foundation for all future modules. • Scaled the Team from 0 → 1: Played a pivotal role in growing the AI and backend teams—designing interviews, onboarding other engineers, mentoring interns, and guiding contributors across codebases and product domains. • Cross-Functional Glue: Bridged AI with backend, UI, and ops—leading delivery of real-time assistants, Slackbots, and demos used in high-stakes board-level presentations and early client trials. • Demo-to-Product Builder: Turned scrappy prototypes into business-critical demos that unlocked client/ board confidence. Co-led the first assistant showcase, helping define the product’s trajectory and value story
SliceX AI
Jun 20229 mosFeb 2023

San Francisco Bay Area

- Spearheaded the ideation and execution of a project to revamp a baseline research paper search engine, integrating it into a Slack-bot and utilizing SliceX AI models similar to GPT-3. Resulted in an improved search efficiency that enhanced productivity of academic research across 2.2 million records of papers that the model was trained on. - Deployed multiple NLP-based APIs for SliceX AI's public cloud platform, including a financial customer support ticket classifier. Contributed to the automation of customer support processes, enabling faster and more efficient ticket classification. - Represented the company at the Edge Computing World Conference 2022, liaising with clients to better understand their needs. Utilized the conference as a platform to introduce and gather feedback for SliceX AI's product offerings, leading to an increase in client satisfaction and better product-market fit.
MindMics
Sep 202110 mosJun 2022

Cambridge, Massachusetts, United States

- Led a research and development project aimed at innovating new ML algorithms to detect states of the Autonomous Nervous System, as part of the R&D team, with a focus on providing real-time biofeedback to users using the Mindmics device with an ECG-like signal. Resulted in the development of cutting-edge technology that improves the understanding of physiological signals and enables real-time biofeedback to users. - Designed a proof-of-concept for a user-centered dashboard with a real-time visualization of the validated states of the Autonomous Nervous System, which was then integrated into the product by the UI team. Contributed to the development of a user-friendly and informative dashboard, enhancing the user experience. - Worked towards creating new metrics from the wearable sensor data, comparing it with benchmarked ECG signals, improving the accuracy of physiological data analysis. Contributed to the improvement of data analysis methods, resulting in more accurate and reliable physiological data analysis. - Designed and conducted in-house trials to explore and validate new metrics to understand physiological sigh and its effects on the ANS, leveraging both quantitative and qualitative data to inform further algorithm development. Resulted in the development of new metrics to understand physiological signals, improving the accuracy and effectiveness of the technology.

Revanth’s AI Stack

hugging face
hugging face
langchain
langchain
ollama
ollama
openai
openai

Education

Carnegie Mellon University

Master of Science - MS, Electrical and Computer Engineering

Carnegie Mellon University

City University of Hong Kong

Bachelor of Honors in Engineering (BE) with First Class Honors, Electronics and Communications Engineering

City University of Hong Kong

Join Revanth and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Revanth Banala on Tal

Headline

Founder, Lumere AI | Building neuroscience-informed adaptive learning | CMU

Location

Bengaluru

AI stack

hugging face

langchain

ollama

openai

Work experience

Founder at Lumere AI, Nov 2025 - Present · 8 mos, Bengaluru, AI for education — we help institutions understand how their students actually learn. Our engine detects specific misconceptions, tracks them across sessions, and gives teachers real-time visibility into who needs help and on what. Not "Rahul scored 34%." More like "Rahul has been confusing the same concept for 3 weeks and nobody's caught it." Started with JEE/NEET. Building toward K-12. Live product. Serving institutions. That's what we're building. For institutions. For students. At scale.

Chief Technology Officer at Coldbean.ai, Aug 2024 - Aug 2025 · 1 yr 1 mo, San Francisco Bay Area, Joined as an AI Strategy Advisor and later took on the full-time CTO role (while at UnifyCx), architecting Coldbean’s AI stack end-to-end. Built agentic systems for sales intelligence, enrichment, outreach, and onboarding. Scaled the tech team from zero, drove IP-grade innovation, and shipped multiple production-ready AI workflows. The journey eventually closed, but it cemented my conviction for solo-founder-led, execution-driven AI products — a philosophy now core to my next venture.

Senior AI Research Engineer - AI Product at UnifyCX, Mar 2024 - Aug 2025 · 1 yr 6 mos, Santo Domingo, Dominican Republic, Led critical initiatives across platform architecture, LLM-based agent design, and enterprise demos—blending deep technical execution with product strategy and research-backed innovation. My role spanned AI research, infra, internal tooling, and customer-facing solutions. Key Contributions: • Inventor of a Patent-Worthy Intent Platform: Single-handedly built a scalable alternative to systems like Amazon Lex—initially a scrappy side project with a group of interns. Now used across all deployments, it cut onboarding time from ~7 days to 20 minutes and is under patent consideration. • LLM Agent Architecture & Production Systems: Refactored core flows into modular, agentic systems capable of dynamic routing, follow-ups, and summarization. Drove platformization efforts to enable reuse across bots, backend services, and client-facing AI agents. • Enterprise-Grade PoC that lead to a $30M+ engagement: Designed and led delivery of a production-ready demo that directly contributed to a $30M+ enterprise engagement. The solution still anchors high-stakes executive pitches and live sales use cases. • Feature Lead for Study Buddy: Owned end-to-end ideation, design, and execution of a contact-center agent support tool with high engagement. Shaped AI capabilities across multiple agent-assist modules through hands-on delivery and technical strategy. • Product Co-Creation & Insight Mining: Worked closely with Product to define roadmap for features like Voice of Customer. Delivered insights from AI-driven mining that surfaced user pain points product teams hadn’t uncovered—earning praise from stakeholders. Beyond these, I consistently pushed platform thinking across teams—building internal tools like model registries, reusable config runners, agentic flows —accelerating AI deployment velocity org-wide.

Founding AI Engineer at UnifyCX, Mar 2023 - Mar 2024 · 1 yr 1 mo, Boston, Massachusetts, United States, Joined as the first AI hire and helped shape the foundation of Lightbird’s GenAI platform—driving hands-on development, research, and early team growth while wearing multiple hats across engineering, product, and client success. Key Contributions: • Built the First Wave of AI Modules: Designed, prototyped, and productionized critical systems—ticket categorization, intent recognition, QnA systems, product detection, and query enrichment—that still anchor the AI assistant today. • Hands-On Technical Leadership: Drove core AI infra from scratch—defining DTO standards, test frameworks, deployment templates, and prompt stacks—used as the foundation for all future modules. • Scaled the Team from 0 → 1: Played a pivotal role in growing the AI and backend teams—designing interviews, onboarding other engineers, mentoring interns, and guiding contributors across codebases and product domains. • Cross-Functional Glue: Bridged AI with backend, UI, and ops—leading delivery of real-time assistants, Slackbots, and demos used in high-stakes board-level presentations and early client trials. • Demo-to-Product Builder: Turned scrappy prototypes into business-critical demos that unlocked client/ board confidence. Co-led the first assistant showcase, helping define the product’s trajectory and value story

Software Engineer - Machine Learning at SliceX AI, Jun 2022 - Feb 2023 · 9 mos, San Francisco Bay Area, - Spearheaded the ideation and execution of a project to revamp a baseline research paper search engine, integrating it into a Slack-bot and utilizing SliceX AI models similar to GPT-3. Resulted in an improved search efficiency that enhanced productivity of academic research across 2.2 million records of papers that the model was trained on. - Deployed multiple NLP-based APIs for SliceX AI's public cloud platform, including a financial customer support ticket classifier. Contributed to the automation of customer support processes, enabling faster and more efficient ticket classification. - Represented the company at the Edge Computing World Conference 2022, liaising with clients to better understand their needs. Utilized the conference as a platform to introduce and gather feedback for SliceX AI's product offerings, leading to an increase in client satisfaction and better product-market fit.

R&D Data Scientist at MindMics, Sep 2021 - Jun 2022 · 10 mos, Cambridge, Massachusetts, United States, - Led a research and development project aimed at innovating new ML algorithms to detect states of the Autonomous Nervous System, as part of the R&D team, with a focus on providing real-time biofeedback to users using the Mindmics device with an ECG-like signal. Resulted in the development of cutting-edge technology that improves the understanding of physiological signals and enables real-time biofeedback to users. - Designed a proof-of-concept for a user-centered dashboard with a real-time visualization of the validated states of the Autonomous Nervous System, which was then integrated into the product by the UI team. Contributed to the development of a user-friendly and informative dashboard, enhancing the user experience. - Worked towards creating new metrics from the wearable sensor data, comparing it with benchmarked ECG signals, improving the accuracy of physiological data analysis. Contributed to the improvement of data analysis methods, resulting in more accurate and reliable physiological data analysis. - Designed and conducted in-house trials to explore and validate new metrics to understand physiological sigh and its effects on the ANS, leveraging both quantitative and qualitative data to inform further algorithm development. Resulted in the development of new metrics to understand physiological signals, improving the accuracy and effectiveness of the technology.

Education

Master of Science - MS, Electrical and Computer Engineering, Carnegie Mellon University

Bachelor of Honors in Engineering (BE) with First Class Honors, Electronics and Communications Engineering, City University of Hong Kong