Help & Supportcareers@gravityer.com
AI Engineer with 1+ years in RAG systems & multimodal search.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Engineer with production experience in RAG systems, multimodal search, and end-to-end ML pipelines. Skilled in LLM integration, vector databases, NLP, and financial ML with a strong mathematical foundation from IIT Hyderabad. Passionate about building intelligent systems across LLMs, Small Language Models (SLMs), generative AI, multimodal AI, and quantitative FinTech applications including stock market analysis and mutual fund prediction.
IIT Hyderabad
B.Tech · Engineering Physics
August 1, 2020 – June 30, 2024
Alpes.AI
AI Engineer
November 1, 2024 – May 1, 2026
Hyderābād, Telangana, India
Legal Case Search Engine - Supreme Court of India
June 1, 2026 – Present
Trained a retrieval + summarisation model on 2,000 divorce case judgments (160,000 paragraphs), treating each paragraph as a semantic knowledge unit. Embedded paragraphs using HuggingFace sentence-transformers and paraphrase-multilingual models, enabling cross-lingual semantic retrieval. Delivered natural-language query interface that retrieves the most relevant judgment paragraph and auto-generates a case summary — reducing legal research time significantly.
Multimodal Search System
June 1, 2026 – Present
Built a cross-modal search engine supporting text-to-image and image-to-image retrieval using OpenCLIP to generate joint visual-text embeddings in a shared vector space. Benchmarked FAISS indexing strategies for multimodal retrieval across e-commerce, media archive, and legal datasets. Demonstrated vision-language model integration applicable to product search, document understanding, and generative AI pipelines.
Mutual Fund Return Prediction & Recommendation System
June 1, 2026 – Present
Built an end-to-end ML pipeline on historical Indian mutual fund data using Linear Regression, Random Forest, and XGBoost achieved strongest out-of-sample generalisation. Developed a goal-based financial planning module: ingests user risk appetite and investment goals (wealth generation, retirement, child education) and recommends aligned mutual funds. Applied feature engineering on time-series NAV data; system design is extensible to stock return prediction and portfolio optimisation use cases.
Video Frame Prediction Model
June 1, 2026 – Present
Trained a sequential ML model to predict future frames in a video stream using temporal dependencies — applicable to surveillance automation and anomaly detection.
Movie Recommendation System
August 1, 2023 – November 1, 2023
Built a content-based recommendation engine using TF-IDF vectorisation and cosine similarity — foundational technique applicable to personalised financial product and content recommendation.
Audio Classification using ML
January 1, 2023 – March 1, 2023
Classified audio into speech, music, and environmental sounds using ANN and LSTM; extracted MFCC features with Librosa — pipeline design directly applicable to ASR/STT systems. Evaluated with accuracy and cross-entropy loss; optimised with Adam — achieving strong separation across audio categories.
Cultural Fit Analysis
The candidate's diverse project portfolio, spanning academic and professional contexts, and across various AI sub-fields (NLP, computer vision, speech AI, FinTech AI), indicates a broad interest and adaptability. The experience at Alpes.AI and the academic background from IIT Hyderabad suggest a drive for innovation and a solid technical foundation, aligning well with a culture that values technical excellence and continuous learning. The leadership roles in sports also point to a team-oriented mindset.
Soft Skills & Operational Fit
The candidate's project descriptions and work experience indicate a strong problem-solving aptitude and a results-oriented approach, particularly in optimizing AI models and pipelines. The leadership and extracurricular activities (hockey captain, athletics medals) suggest discipline, teamwork, and the ability to perform under pressure, which are valuable for operational fit. However, without direct assessment data on communication or collaboration in a professional setting, this remains an inference.